2
3
4
5
9
10
11
12
United States
Environmental Protection
Agency
Office of Water
Office of Science and Technology
Washington, DC 20460
EPA-823-B-05-
003
December 2006
www.epa.gov
Draft Nutrient Criteria
Technical Guidance Manual
Wetlands
13
-------
15 DISCLAIMER
16
17
18 This manual provides technical guidance to States authorized Tribes, and other authorized
19 jurisdictions to establish water quality criteria and standards under the Clean Water Act (CWA),
20 in order to protect aquatic life from acute and chronic effects of nutrient overenrichment. Under
21 the CWA, States and authorized Tribes are directed to establish water quality criteria to protect
22 designated uses. States and authorized Tribes may use approaches for establishing water quality
23 criteria that differ from the approaches recommended in this guidance. This manual constitutes
24 EPA's scientific recommendations regarding the development of numeric criteria reflecting
25 ambient concentrations of nutrients that protect aquatic life. However, it does not substitute for
26 the CWA or EPA's regulations; nor is it a regulation itself. Thus, it cannot impose legally
27 binding requirements on EPA, States, Authorized Tribes, or the regulated community, and might
28 not apply to a particular situation or circumstance. Further, States and Authorized Tribes may
29 choose to develop different types of nutrient criteria for wetlands that are scientifically
30 defensible and protective of the designated use, including narrative criteria. EPA may change
31 this guidance in the future.
-------
32
33
34
35
36
37
38
39
40
41
42
43
44 THIS PAGE INTENTIONALLY
45 LEFT BLANK
-------
December 2006 DRAFT
46 TABLE OF CONTENTS
47
48 CONTRIBUTORS vii
49 ACKNOWLEDGMENTS viii
50 LIST OF FIGURES ix
51 LIST OF TABLES x
52 LIST OF INTERNET LINKS/REFERENCES xi
53 EXECUTIVE SUMMARY 1
54 Chapter 1 Introduction 5
55 1.1 INTRODUCTION 5
56 1.2 WATER QUALITY STANDARDS AND CRITERIA 6
57 1.3 NUTRIENT ENRICHMENT PROBLEMS 8
58 1.4 OVERVIEW OF THE CRITERIA DEVELOPMENT PROCESS 11
59 1.5 ROADMAP TO THE DOCUMENT 12
60 Chapter 2 Overview of Wetland Science 16
61 2.1 INTRODUCTION 16
62 2.2 COMPONENTS OF WETLANDS 17
63 2.3 WETLAND NUTRIENT COMPONENTS 22
64 Chapter 3 Classification of Wetlands 30
65 3.1 INTRODUCTION 30
66 3.2 EXISTING WETLAND CLASSIFICATION SCHEMES 31
67 3.3 SOURCES OF INFORMATION FOR MAPPING WETLAND CLASSES 42
68 3.4 DIFFERENCES IN NUTRIENT REFERENCE CONDITION OR SENSITIVITY TO NUTRIENTS
69 AMONG WETLAND CLASSES 45
70 3.5 RECOMMENDATIONS 46
71 Chapter 4 Sampling Design for Wetland Monitoring 50
72 4.1 INTRODUCTION 50
73 4.2 CONSIDERATIONS FOR SAMPLING DESIGN 52
74 4.3 SAMPLING PROTOCOL 57
75 4.4 SUMMARY 63
76 Chapter 5 Candidate Variables for Establishing Nutrient Criteria 65
77 5.1 OVERVIEW OF CANDIDATE VARIABLES 65
78 5.2 SUPPORTING VARIABLES 67
79 5.3 CAUSAL VARIABLES 70
80 5.4 RESPONSE VARIABLES 76
81 Chapter 6 Database Development and New Data Collection 82
82 6.1 INTRODUCTION 82
83 6.2 DATABASES AND DATABASE MANAGEMENT 82
84 6.3 QUALITY OF HISTORICAL AND COLLECTED DATA 86
85 6.4 COLLECTING NEW DATA 88
86 6.5 QUALITY ASSURANCE / QUALITY CONTROL (QA/QC) 91
87 Chapter 7 Data Analysis 92
88 7.1 INTRODUCTION 92
89 7.2 FACTORS AFFECTING ANALYSIS APPROACH 92
90 7.3 DISTRIBUTION-BASED APPROACHES 94
91 7.4 RESPONSE-BASED APPROACHES 95
92 7.5 PARTITIONING EFFECTS AMONG MULTIPLE STRESSORS 97
93 7.6 STATISTICAL TECHNIQUES 98
94 7.7 LINKING NUTRIENT AVAILABILITY TO PRIMARY PRODUCER RESPONSE 101
95 Chapter 8 Criteria Development 104
IV
-------
December 2006 DRAFT
96 8.1 INTRODUCTION 104
97 8.2 METHODS FOR DEVELOPING NUTRIENT CRITERIA 105
98 8.4 INTERPRETING AND APPLYING CRITERIA 112
99 8.5 SAMPLING FOR COMPARISON TO CRITERIA 112
100 8.6 CRITERIA MODIFICATIONS 113
101 REFERENCES 114
102 APPENDIX A. ACRONYM LIST AND GLOSSARY 139
103 ACRONYMS 139
104 GLOSSARY 141
105 APPENDIX B. CASE STUDY 1: DERIVING A PHOSPHORUS CRITERION FOR THE FLORIDA
106 EVERGLADES 145
107 APPENDIX B. CASE STUDY 2: THE BENEFICIAL USE OF NUTRIENTS FROM TREATED
108 WASTEWATER EFFLUENT IN LOUISIANA WETLANDS: A REVIEW 170
-------
109
110
111
112
113
114
115
116
117
118
119 THIS PAGE INTENTIONALLY
120 LEFT BLANK
121
VI
-------
December 2006 DRAFT
122
123
124 CONTRIBUTORS
125
126
127 Nancy Andrews (U.S. Environmental Protection Agency)
128 Mark Clark (University of Florida)
129 Christopher Craft (University of Indiana)
130 William Crumpton (Iowa State University)
131 Ifeyinwa Davis (U.S. Environmental Protection Agency)
132 Naomi Detenbeck (U.S. Environmental Protection Agency)
133 Paul McCormick (U.S. Geological Survey)
134 Amanda Parker (U.S. Environmental Protection Agency)*
135 Kristine Pintado (Louisiana Department of Environmental Quality)
136 Steve Potts (U.S. Environmental Protection Agency)
137 Todd Rasmussen (University of Georgia)
138 Ramesh Reddy (University of Florida)
139 R. Jan Stevenson (Michigan State University)
140 Arnold van der Valk (Iowa State University)
141
142
143 *Principal Author
Vll
-------
December 2006 DRAFT
144
145
146 ACKNOWLEDGMENTS
147
148
149 The authors wish to acknowledge the efforts and input of several individuals. These include
150 members of our EPA National Nutrient Team: Jim Carleton, Lisa Larimer and Sharon Frey;
151 members of the EPA Wetlands Division: Kathy Hurld, Chris Faulkner and Donna Downing; and
152 members of the Office of General Counsel: Leslie Darman and Paul Bangser. We also want to
153 thank Kristine Pintado (DEQ, Louisiana) for her contributions and her careful review and
154 comments.
155
156 This document was peer reviewed by a panel of expert scientists. The peer review charge
157 focused on evaluating the scientific validity of the processes and techniques for developing
158 nutrient criteria described in the guidance. The peer review panel comprised Dr. Russel B.
159 Frydenborg, Dr. Robert H. Kadlec, Dr. Lawrence Richards Pomeroy, Dr. Eliska Rejmankova and
160 Dr. Li Zhang. Edits and suggestions made by the peer review panel were incorporated into the
161 final version of the guidance.
Vlll
-------
December 2006 DRAFT
162
163 LIST OF FIGURES
164
165 Figure 1.1 Flowchart providing the steps of the process to develop wetland nutrient
166 criteria
167 Figure 2.1 Schematic of nutrient transfer among potential system sources and sinks.
168 Figure 2.2 Relationship between water source and wetland vegetation. Modified from
169 Brinson(1993)
170 Figure 2.3 Schematic showing basic nutrient cycles in soil-water column of a wetland
171 Figure 2.4 Range of redox potentials in wetland soils
172 Figure 2.5 Schematic of the nitrogen cycle in wetlands
173 Figure 2.6 Schematic of the phosphorus cycle in wetlands
174 Figure 3.1 Map of Omernik aquatic ecoregions
175 Figure 3.2 Map of Bailey ecoregions with coastal and estuarine provinces
176 Figure 3.3 Examples of first four hierarchical levels of Ecological Units
177 Figure 3.4 Dominant water sources to wetlands, from Brinson 1993
178 Figure 3.5 Dominant hydrodynamic regimes for wetlands based on flow pattern
179 Figure 3.6 Interaction with break in slope with groundwater inputs to slope wetlands
180 Figure 3.7 (Top) Cowardin hierarchy of habitat types for estuarine systems
181 (Bottom) Palustrine systems, from Cowardin et al. 1979
182 Figure 5.1 Conceptual model of causal pathway between human activities and ecological
183 attributes.
184 Figure 7.1 Biological condition gradient model describing biotic community condition as
185 levels of stressors increase.
186 Figure 8.1 Use of undisturbed wetlands as a reference for establishing criteria versus an
187 effects based approach.
188 Figure 8.2 Tiered aquatic life use model used in Maine.
189 Figure 8.3 Percent calcareous algal mat cover in relation to distance from the P source
190 showing the loss of the calcareous algal mat in those sites closer to the source
IX
-------
December 2006 DRAFT
191
192 LIST OF TABLES
193
194 Table 1 Observed consequences of cultural eutrophication in freshwater wetlands
195
196 Table 2 Comparison of landscape and wetland classification schemes
197
198 Table 3 Features of the major hydrogeomorphic classes of wetlands that may
199 influence background nutrient concentrations, sensitivity to nutrient loading,
200 nutrient storage forms and assimilative capacity, designated use and choice
201 ofendpoints
202
203 Table 4 Comparison of Stratified Probabilistic, Targeted, and BACI Sampling Designs
-------
December 2006 DRAFT
204
205
206
207 Chapter 1
208
209
210
211
212 Chapter 2
213
214
215 Chapter 3
216
217
218
219
220
221
222
223
224
225 Chapter 4
226
227
228
229
230
231
232
233
234
235
236 Chapter 5
237
238
239
240
241 Chapter 6
242
243
244
245 Chapter 7
246
LIST OF INTERNET LINKS/REFERENCES
http ://www. epa. sov/waterscience/criteria/nutrient/suidance/index.html
http ://www. epa. sov/owow/wetlands
http://www.epa.sov/waterscience/criteria/nutrient/stratesv.html.
http ://www. epa. sov/owow/wetlands/initiative
http://www.arl.noaa.sov/research/prosrams/airmon.html
http://nadp.sws.uiuc.edu/
http ://www. epa. sov/emap/remap/index.html
http ://www. epa. sov/bioindicators/
http ://www. epa. sov/waterscience/standards/nutrient.html
http://el.erdc.usace.armv.mil/wetlands/pdfs/wrpde9.pdf
http://water.usss.sov/GIS/metadata/ussswrd/XML/ecoresion.xml
http ://www.nwi .fws. sov
http://www.wes.armv.mil/el/wetlands
http://www.epa.sov/waterscience/criteria/wetlands/7Classification.pdf
http://www.epa.sov/waterscience/criteria/wetlands/17LandUse.pdf
http ://www. epa. sov/waterscience/criteri a/wetlands/index. html
http ://www. epa. sov/owow/wetlands/bawws/case/me.html
http ://www. epa. sov/owow/wetlands/bawws/case/mtdev.html
http ://www. epa. sov/owow/wetlands/bawws/case/wa.html
http ://www. epa. sov/owow/wetlands/bawws/case/fl 1 .html
http ://www. epa. sov/owow/wetlands/bawws/case/fl2.html
http ://www. epa. sov/owow/wetlands/bawws/case/oh 1 .html
http ://www. epa. sov/owow/wetlands/bawws/case/mn 1 .html
http ://www. epa. sov/owow/wetlands/bawws/case/or.html
http://www.epa.sov/owow/wetlands/bawws/case/wi 1 .html
http://www.epa.sov/waterscience/criteria/wetlands/10Vesetation.pdf
http ://www. epa. sov/waterscience/criteri a/wetlands/ 1 1 Alsae.pdf
http://www.epa.sov/waterscience/criteria/wetlands/9Invertebrate.pdf
http://www.epa.sov/waterscience/criteria/wetlands/16Indicators.pdf
http://www.nps.ars.usda.sov/prosrams/nrsas.htm
http://www.fs.fed.us/research/
http://www.usbr.sov
http://el.erdc.usace.armv.mil/wrap/wrap.html
http://firehole.humboldt.edu/wetland/twdb.html
XI
-------
December 2006 DRAFT
247 http://www.socialresearchtnethods.net/
248 http://www.math.yorku.ca/SCS/
249 http://calculators.stat.ucla.edu/powercalc/
250 http://www.survevsystem.com/sscalc.htm
251 http://www.health.ucalgary.ca/~rollin/stats/ssize/index.html
252 http ://www. stat. ohi o-state. edu/~j ch/ssinput.html
253 http ://www. stat.uiowa. edu
254 http ://www. epa. gov/waterscience/criteria/wetlands
255
256 Chapter 8 http://www.epa.gov/waterscience/biocriteria/modules/wetl01-05-alus-
257 monitoring.pdf
258 http ://www. epa. gov/owow
259 http://www.wetlandbiogeochemistry.lsu.edu/
260 http://data.lca.gov/Ivan6/app/app c ch9.pdf
Xll
-------
December 2006 DRAFT Executive Summary
261
262
263 EXECUTIVE SUMMARY
264
265 The purpose of this document is to provide scientifically defensible guidance to assist States,
266 Tribes, and Territories in assessing the nutrient status of their wetlands, and to provide technical
267 assistance for developing regionally-based numeric nutrient criteria for wetland systems. The
268 development of nutrient criteria is part of an initiative by the US Environmental Protection
269 Agency (USEPA) to address the problem of cultural eutrophication, i.e., excess nutrients caused
270 by human activities (USEPA 1998a). Cultural eutrophication is not new; however, traditional
271 efforts at nutrient control have been only moderately successful. Specifically, efforts to control
272 nutrients in water bodies that have multiple nutrient sources (point and nonpoint sources) have
273 been less effective in providing satisfactory, timely remedies for enrichment-related problems.
274 The development of numeric criteria should aid control efforts by providing clear numeric goals
275 for nutrient concentrations. Furthermore, numeric nutrient criteria provide specific water quality
276 goals that will assist researchers in designing improved best management practices.
277
278 In 1998, the USEPA published a report entitled National Strategy for the Development of
279 Regional Nutrient Criteria (USEPA 1998a). This report outlines a framework for development
280 of waterbody-specific technical guidance that can be used to assess nutrient status and develop
281 region-specific numeric nutrient criteria. The document presented here is the wetland-specific
282 technical guidance for developing numeric nutrient criteria. The Nutrient Criteria Technical
283 Guidance Manuals for Rivers and Streams (USEPA, 2000b), Lakes and Reservoirs (USEPA,
284 2000a) and Estuarine and Coastal Marine Waters (USEPA, 2001) have been completed and are
285 available at: http://www.epa.gov/waterscience/criteria/nutrient/guidance/index.html.
286 Section 303(c) of the Clean Water Act directs states to adopt water quality standards for
287 interstate and intrastate waters that are "waters of the United States". Wetlands are
288 included in the definition of "waters of the United States" (40 C.F.R. 230.2(s)). States
289 should therefore have water quality criteria to protect the designated uses of wetlands that
290 are waters of the U.S. in addition to other surface water types (lakes, streams, estuaries)
291 that have traditionally been monitored and regulated for water quality. This guidance is to
292 assist states in developing numeric nutrient criteria for wetlands, should the State or
293 Autorized Tribe choose to do so. Further, States and Authorized Tribes may choose to
294 develop different types of criteria for wetlands protection, including narrative criteria.
295
296 In this document, the term waterbody is used generically to encompass a wide range of aquatic
297 habitats, from lentic and lotic systems with permanent standing water to wetland systems that
298 have saturated sediments but no standing water, or which are flooded only temporarily. EPA
299 recommends that States, Territories and Tribes' include wetlands in the water quality standards
300 definition of "State waters" by adopting a regulatory definition of "State waters" at least as
301 inclusive as the Federal definition of "waters of the U.S.", and adopting an appropriate definition
302 for "wetlands".
1
-------
December 2006 DRAFT Executive Summary
303
304 CLASSIFICATION OF WETLANDS
305
306 Classification strategies for nutrient criteria development include:
307 • physiographic regions
308 • hydrogeomorphic class
309 • water depth and duration
310 • vegetation type or zone
311 Choosing a specific classification scheme will likely depend on practical considerations, such as:
312 whether a classification scheme is available in mapped digital form or can be readily derived
313 from existing map layers; whether a hydrogeomorphic or other classification scheme has been
314 refined for a particular region and wetland type; and whether classification schemes are already
315 in use for monitoring and assessment of other waterbody types in a state or region.
316
31? SAMPLING DESIGN
318
319 Three sampling designs for new wetland monitoring programs are described:
320 • probabilistic sampling
321 • targeted/tiered approach
322 • BACI (Before/After, Control/Impact)
323 These approaches are designed to obtain a significant amount of information for statistical
324 analyses with relatively minimal effort. Sampling efforts should be designed to collect
325 information that will answer management questions in a way that will allow robust statistical
326 analysis. In addition, site selection, characterization of reference sites or systems, and
327 identification of appropriate index periods are all of particular concern when selecting an
328 appropriate sampling design. Careful selection of sampling design will allow the best use of
329 financial resources and will result in the collection of high quality data for evaluation of the
330 wetland resources of a State or Tribe.
331
332 CANDIDATE VARIABLES FOR ESTABLISHING NUTRIENT CRITERIA
333
334 Candidate variables to use in determining nutrient condition of wetlands and to help identify
335 appropriate nutrient criteria for wetlands consist of supporting variables, causal variables, and
336 response variables. Supporting variables provide information useful in normalizing causal and
337 response variables and categorizing wetlands. Causal variables are intended to characterize
338 nutrient availability (or assimilation) in wetlands and could include nutrient loading rates and
339 soil nutrient concentrations. Response variables are intended to characterize biotic response and
-------
December 2006 DRAFT Executive Summary
340 could include community structure and composition of macrophytes and algae. Recommended
341 variables for wetland nutrient criteria development described in this chapter are:
342 1. Causal variables - nutrient loading rates, land use, extractable and total soil nitrogen (N) and
343 phosphorus (P), water column N and P;
344 2. Response variables - nutrient content of wetland vegetation (algal and/or higher plants),
345 aboveground biomass and stem height, macrophyte, algal, and macroinvertebrate community
346 structure and composition;
347 3. Supporting variables - hydrologic condition/balance, conductivity, soil pH, soil bulk density,
348 particle size distribution, soil organic matter content.
349 DATABASE DEVELOPMENT AND NEW DATA COLLECTION
350 A database of relevant water quality information can be an invaluable tool to States and Tribes as
351 they develop nutrient criteria. In some cases existing data are available and can provide
352 additional information that is specific to the region where criteria are to be set. However, little or
353 no data are available for most regions or parameters, and creating a database of newly gathered
354 data is strongly recommended. In the case of existing data, the data should be geolocated, and
355 their suitability (type and quality and sufficient associated metadata) ascertained.
356 DATA ANALYSIS
357 Data analysis is critical to nutrient criteria development. Proper analysis and interpretation of
358 data determine the scientific defensibility and effectiveness of the criteria. Therefore, it is
359 important to evaluate short and long-term goals for wetlands of a given class within the region of
360 concern. The purpose of this chapter is to explore methods for analyzing data that can be used to
361 develop nutrient criteria consistent with these goals. Techniques discussed in this chapter
362 include:
363 • Distribution based approaches that examine distributions of primary and supporting
364 variables (i.e., the percentile approach);
365 • Response based approaches that develop relationships between measurements of nutrient
366 exposure and ecological responses (i.e., tiered aquatic life uses);
367 • Partitioning effects of multiple stressors;
368 • Statistical techniques;
369 • Multi-metric indices;
370 • Linking nutrient availability to primary producer response.
371
372 CRITERIA DEVELOPMENT
-------
December 2006 DRAFT Executive Summary
373
374 Several methods can be used to develop numeric nutrient criteria for wetlands; they include, but
375 are not limited to, criteria development methods that are detailed in this document:
376
377 • Comparing conditions in known reference systems for each established
378 wetland type and class based on using best professional judgment (BPJ) or
379 identifying reference conditions using frequency distributions of empirical
380 data and identifying criteria using percentile selections of data plotted as
381 frequency distributions.
382 • Refining classification systems, using models, and/or examining system
383 biological attributes in comparison to known reference conditions to
384 assess the relationships among nutrients, vegetation or algae, soil, and
385 other variables and identifying criteria based on thresholds where those
386 response relationships change.
387 • Using or modifying published nutrient and vegetation, algal, and soil
388 relationships and values to identify appropriate criteria.
389
390 A weight of evidence approach with multiple attributes that combine one or more of the
391 development approaches will produce criteria of greater scientific validity.
392
393 Once criteria are developed, they should be implemented into state water quality programs to be
394 effective. The implementation procedures, particularly for wetland systems, may be complex and
395 will likely vary greatly from state to state. The purpose of this document is to provide guidance
396 on developing numeric nutrient criteria in a scientifically valid manner, and is not intended to
397 address the multiple, complex issues surrounding implementation of water quality criteria and
398 standards. Implementation will be addressed in a different process and additional implementation
399 assistance will also be provided through other technical assistance projects provided by EPA.
400 For issues specific to constructed wetlands, States and Tribes should refer to
401 http ://www. epa. gov/owow/wetlands/watersheds/cwetl ands.html.
-------
December 2006 DRAFT Chapter 1. Introduction
402 Chapter 1 Introduction
403
404 1.1 INTRODUCTION
405
406 PURPOSE
407
408 The purpose of this document is to provide technical guidance to assist States and Tribes in
409 assessing the nutrient status of their wetlands by considering water, vegetation and soil
410 conditions, and to provide technical assistance for developing regionally-based, scientifically
411 defensible, numeric nutrient criteria for wetland systems.
412
413 EPA's development of recommended nutrient criteria is part of an initiative by the US
414 Environmental Protection Agency (USEPA) to address the problem of cultural eutrophication. In
415 1998, the EPA published a report entitled National Strategy for the Development of Regional
416 Nutrient Criteria (USEPA 1998a). The report outlines a framework for development of
417 waterbody-specific technical guidance that can be used to assess nutrient status and develop
418 region-specific numeric nutrient criteria. This document is the technical guidance for developing
419 numeric nutrient criteria for wetlands. Additional more specific information on sampling
420 wetlands is available at: http://www.epa.gov/waterscience/criteria/nutrient/guidance/.
421
422
423 AUTHORITY
424
425 Section 303(c) of the Clean Water Act directs states to adopt water quality standards for
426 interstate and intrastate waters that are "waters of the United States". Wetlands are
427 included in the definition of "waters of the United States" (40 C.F.R. 230.2(s)).
428
429 In this document, the term waterbody is used generically to encompass a wide range of aquatic
430 habitats, from lentic and lotic systems with permanent standing water to wetland systems that
431 have saturated sediments but no standing water, or which are flooded only temporarily. Wetlands
432 must be legally included in the scope of States' and Tribes' water quality standards programs for
433 water quality standards to be applicable to wetlands. EPA recommends that States and Tribes
434 include wetlands in the water quality standards definition of "State waters" by adopting a
435 regulatory definition of "State waters" at least as inclusive as the Federal definition of "waters of
436 the U.S.", and adopting an appropriate definition for "wetlands". Examples of different state
437 approaches can be found at: http://www.epa.gov/owow/wetlands/initiatives/.
438
439 Discussions about water quality in this document refer to wetland systems as waters of the US
440 under the authority given to the USEPA in the CWA. EPA recognizes that wetland systems are
441 different from the other waters of the US in that they frequently do not have standing or flowing
442 water, and that the soils and vegetation components are more dominant in these systems than in
443 the other waterbody types (lakes, streams, estuaries).
-------
December 2006 DRAFT Chapter 1. Introduction
444
445 BACKGROUND
446
447 Cultural eutrophication (human-caused inputs of excess nutrients in waterbodies) is one of the
448 primary factors resulting in impairment of surface waters in the US (USEPA 1998a). Both point
449 and nonpoint sources of nutrients contribute to impairment of water quality. Point source
450 discharges of nutrients are relatively constant and are controlled by the National Pollutant
451 Discharge Elimination System (NPDES) permitting program. Nonpoint pollutant inputs have
452 increased in recent decades resulting in degraded water quality in many aquatic systems.
453 Nonpoint sources of nutrients are most commonly intermittent and are usually linked to runoff,
454 atmospheric deposition, seasonal agricultural activity, and other irregularly occurring events
455 such as silvicultural activities. Control of nonpoint source pollutants typically focuses on land
456 management activities and regulation of pollutants released to the atmosphere.
457
458 The term eutrophi cation was coined in reference to lake systems. The use of the term for other
459 waterbody types can be problematic due to the confounding nature of hydrodynamics, light, and
460 other waterbody type differences on the responses of algae and vegetation. Eutrophi cation in this
461 document refers to human-caused inputs of excess nutrients and is not intended to indicate the
462 same scale or responses to eutrophi cation found in lake systems and codified in the trophic state
463 index for lakes (Carlson 1977). This manual is intended to provide guidance for identifying
464 deviance from natural conditions with respect to cultural eutrophication in wetland systems.
465 Hydrologic alteration and pollutants other than excess nutrients may amplify or reduce the
466 effects of nutrient pollution, making specific responses to nutrient pollution difficult to quantify.
467 EPA recognizes these issues, and presents recommendations for analyzing wetland systems with
468 respect to nutrient condition for development of nutrient criteria in spite of these confounding
469 factors.
470
471 Cultural eutrophication is not new; however, traditional efforts at nutrient control have been only
472 moderately successful. Specifically, efforts to control nutrients in waterbodies that have multiple
473 nutrient sources (point and nonpoint sources) have been less effective in providing satisfactory,
474 timely remedies for enrichment-related problems. The development of numeric criteria should
475 aid control efforts by providing clear numeric goals for nutrient concentrations. Furthermore,
476 numeric nutrient criteria provide specific water quality goals that will assist researchers in
477 designing improved best management practices.
478
479
480 1.2 WATER QUALITY STANDARDS AND CRITERIA
481
482 States, Territories, and authorized Tribes are responsible for setting water quality standards to
483 protect the physical, biological, and chemical integrity of their waters. "Water quality standards
484 (WQS) are provisions of State or Federal law which consist of a designated use or uses for the
485 waters of the United States and water quality criteria for such waters to protect such uses. Water
-------
December 2006 DRAFT Chapter 1. Introduction
486 quality standards are to protect public health or welfare, enhance the quality of the water, and
487 serve the purposes of the Act (40 CFR 131.3)" (USEPA 1994). A water quality standard defines
488 the goals for a waterbody by: 1) designating its specific uses, 2) setting criteria to protect those
489 uses, and 3) establishing an antidegradation policy to protect existing water quality.
490
491 Water quality criteria may be expressed as numeric or narrative criteria. Most of the Nation's
492 waterbodies do not have numeric nutrient criteria, but instead rely on narrative criteria that
493 describe the desired condition. Narrative criteria are descriptions of conditions necessary for the
494 waterbody to attain its designated. An example of a narrative criterion from Florida is shown
495 below:
496
497 In no case shall nutrient concentrations of a body of water be altered so as to cause
498 an imbalance in natural populations of aquatic flora or fauna.
499
500 Numeric criteria, on the other hand, identify specific values designed to protect specified
501 designated uses such as an aquatic life use. Numeric criteria are values assigned to measurable
502 components of water quality, such as the concentration of a specific constituent that is present in
503 the water column. An example of a numeric criterion is shown below:
504
505 The three month or greater geometric mean of water column total phosphorus [TP]
506 in the Everglades shall not exceed 10 /ug/L.
507
508 In addition to narrative and numeric criteria, some States and Tribes use numeric translator
509 mechanisms—mechanisms that translate narrative (qualitative) standards into numeric
510 (quantitative) values for use in evaluating water quality data—as an intermediate step between
511 numeric criteria and water quality standards that are not written into State or Tribal laws but are
512 used internally by the State or Tribal agency as goals and assessment levels for management
513 purposes.
514
515 Numeric criteria provide distinct interpretations of acceptable and unacceptable conditions, form
516 the foundation for measurement of environmental quality, and reduce ambiguity for management
517 and enforcement decisions. The lack of numeric nutrient criteria for most of the Nation's
518 waterbodies makes it difficult to assess the condition of waters of the US, and to develop
519 protective water quality standards, thus hampering the water quality manager's ability to protect
520 and improve water quality.
521
522 Many States, Tribes, and Territories have adopted some form of nutrient criteria for surface
523 waters related to maintaining natural conditions and avoiding nutrient enrichment. Most States
524 and Tribes with nutrient criteria in their water quality standards have broad narrative criteria for
525 most waterbodies and may also have site-specific numeric criteria for certain waters of the State.
526 Established criteria most commonly pertain to P concentrations in lakes. Nitrogen criteria, where
527 they have been established, are usually protective of human health effects or relate to toxic
528 effects of ammonia and nitrates. In general, levels of nitrate (10 ppm [mg/L] for drinking water)
-------
December 2006 DRAFT Chapter 1. Introduction
529 and ammonia high enough to be problematic for human health or toxic to aquatic life (1.24 mg
530 N/L at pH = 8 and 25°C) will also cause problems of enhanced algal growth (USEPA 1986).
531
532 Numeric nutrient criteria can provide a variety of benefits and may be used in conjunction with
533 State/Tribal and Federal biological assessments, Nonpoint Source Programs, Watershed
534 Implementation Plans, and in development of Total Maximum Daily Loads (TMDLs) to improve
535 resource management and support watershed protection activities at local, State, Tribal, and
536 national levels. Information obtained from compiling existing data and conducting new surveys
537 can provide water quality managers and the public a better perspective on the condition of State,
538 Territorial, and Tribal waters. The compiled waterbody information can be used to most
539 effectively budget personnel and financial resources for the protection and restoration of State
540 waters. In a similar manner, data collected in the criteria development and implementation
541 process can be compared before, during, and after specific management actions. Analyses of
542 these data can determine the response of the waterbody and the effectiveness of management
543 endeavors.
544
545
546 1.3 NUTRIENT ENRICHMENT PROBLEMS
547
548 Water quality can be affected when watersheds are modified by alterations in vegetation,
549 sediment transport, fertilizer use, industrialization, urbanization, or conversion of native forests
550 and grasslands to agriculture and silviculture (Turner and Rabalais 1991; Vitousek et al. 1997;
551 Carpenter et al. 1998). Cultural eutrophication, one of the primary factors resulting in
552 impairment of U. S. surface waters (USEPA 1998a) results from point and nonpoint nutrient
553 sources. Nonpoint pollutant inputs have increased in recent decades and have degraded water
554 quality in many aquatic systems (Carpenter et al. 1998). Control of nonpoint source pollutants
555 focuses on land management activities and regulation of pollutants released to the atmosphere
556 (Carpenter etal. 1998).
557
558 Nutrient enrichment frequently ranks as one of the top causes of water resource impairment. The
559 USEPA reported to Congress that of the waterbodies surveyed and reported impaired, 20 percent
560 of rivers and 50 percent of lakes were listed with nutrients as the primary cause of impairment
561 (USEPA 2000c). Few States or Tribes currently include wetland monitoring in their routine
562 water quality monitoring programs (only eleven States and Tribes reported attainment of
563 designated uses for wetlands in the National Water Quality Inventory 1998 Report to Congress
564 (USEPA 1998b) and only three states used monitoring data as a basis for determining attainment
565 of water quality standards for wetlands); thus, the extent of nutrient enrichment and impairment
566 of wetland systems is largely undocumented. Increased wetland monitoring by States and Tribes
567 will help define the extent of nutrient enrichment problems in wetland systems.
568
569 The best-documented case of cultural eutrophication in wetlands is the Everglades ecosystem.
570 The Everglades ecosystem is a wetland mosaic that is primarily of oligotrophic freshwater
-------
December 2006 DRAFT Chapter 1. Introduction
571 marsh. Historically, the greater Everglades ecosystem included vast acreage of freshwater marsh,
572 stands of custard apple, and some cattail south of Lake Okeechobee and Big Cypress Swamp,
573 that eventually drained into Florida Bay. Lake Okeechobee was diked to reduce flooding. The
574 area directly south of Lake Okeechobee was then converted into agricultural lands for cattle
575 grazing and row crop production. The cultivation and use of commercial fertilizers in the area
576 now known as the "Everglades Agricultural Area" has resulted in release of nutrient-rich waters
577 into the Everglades for more than thirty years. The effects of the nutrient-rich water, combined
578 with coastal development and channeling water to supply water to communities on the southern
579 Florida coast have resulted in significant increases in soil and water column phosphorus levels in
580 naturally oligotrophic areas. In particular, nutrient enrichment of the freshwater marsh has
581 resulted in an imbalance in the native vegetation. Cattail is now encroaching in areas that were
582 historically primarily sawgrass; calcareous algal mats are being replaced by non-calcareous
583 algae, changing the balance of native flora that is needed to support vast quantities of wildlife.
584 Nutrient enriched water is also reaching Florida Bay, suffocating the native turtle grass as
585 periphyton covers the blades (Davis and Ogden 1994; Everglades Interim Report 1999, 2003;
586 Everglades Consolidated Report 2003). Current efforts to restore the Everglades are focusing on
587 nutrient reduction and better hydrologic management (Everglades Consolidated Report 2003).
588
589 Monitoring to establish trends in nutrient levels and associated changes in biology has been
590 infrequent for most wetland types as compared to studies in the Everglades or examination of
591 other surface waters such as lakes. Noe et al. (2001) have argued that phosphorus
592 biogeochemistry and the extreme oligotrophy observed in the Everglades in the absence of
593 anthropogenic inputs represents a unique case. Effects of cultural eutrophication, however, have
594 been documented in a range of different wetland types. Existing studies are available to
595 document potential impacts of anthropogenic nutrient additions to a wide variety of wetland
596 types, including bogs, fens, Great Lakes coastal emergent marshes, and cypress swamps. The
597 evidence of nutrient effects in wetlands ranges from controlled experimental manipulations, to
598 trend or empirical gradient analysis, to anecdotal observations. Consequences of cultural
599 eutrophication have been observed at both community and ecosystem-level scales (Table 1).
600 Deleterious effects of nutrient additions on wetland vegetation composition have been
601 demonstrated in bogs (Kadlec and Bevis 1990), fens (Guesewell et al. 1998, Bollens and
602 Ramseier 2001, Pauli et al. 2002), wet meadows (Finlayson et al. 1986), marshes (Bedford et al.
603 1999) and cypress domes (Ewel 1976). Specific effects on higher trophic levels in marshes seem
604 to depend on trophic structure (e.g., presence/absence of minnows, benthivores, and/or
605 piscivores, Jude and Pappas 1992, Angeler et al. 2003) and timing/frequency of nutrient
606 additions (pulse vs. press; Gabor et al. 1994, Murkin et al. 1994, Hann and Goldsborough 1997,
607 Sandilands et al. 2000, Hann et al. 2001, Zrum and Hann 2002).
608
609 The cycling of nitrogen (N) and phosphorus (P) in aquatic systems should be considered when
610 managing nutrient enrichment. The hydroperiod of wetland systems significantly affects nutrient
611 transformations, availability, transport, and loss of gaseous forms to the atmosphere
-------
December 2006 DRAFT
Chapter 1. Introduction
612
Table 1. Observed consequences of cultural eutrophication in freshwater wetlands.
Observed impact
Loss of submerged aquatic plants that have high
light compensation points
Shifts in vascular plant species composition due to
shifts in competitive advantage
Increases in above-ground production
Decreases in local or regional biodiversity
Increased competitive advantage of
aggressive/invasive species
(e.g., Typhaglauca, T. latifolia and Phalaris
arundinacea)
Loss of nutrient retention capacity (e.g., carbon and
nitrogen storage, changes in plant litter
decomposition)
Major structural shifts between "clear water"
macrophyte dominated systems to turbid
phytoplankton dominated systems or metaphyton-
dominated systems with reduced macrophyte
coverage
Shifts in macroinvertebrate composition along a
cultural eutrophication gradient
References
Phillips et al. 1978
Stephenson et al. 1980
Galatowitsch and van der Valk 1996
Wentz 1976
Verhoeven et al. 1988
Ehrenfeld and Schneider 1993
Gaudet and Keddy 1995
Koerselman and Verhoeven 1995
Barko 1983
Bayley etal. 1985
Barko and Smart 1986
Vermeer 1986
Mudroch and Capobiancol979
Guntenspergen et al. 1980
Lougheed et al. 2001
Balla and Davis 1995
VanGroenendael et al. 1993
Bedford etal. 1999
Woo and Zedler 2002
Svengsouk and Mitsch 2001
Green and Galatowitsch 2002
Maurer and Zedler 2002
Nichols 1983
Davis and van der Valk 1983
Rybczyketal. 1996
McDougal et al. 1997
Angel er et al. 2003
Chessman et al. 2002
613
614
615
616
617
618
619
(Mitsch and Gosselink, 2000). Nutrients can be re-introduced into a waterbody from the
sediment, or by microbial transformation, potentially resulting in a long recovery period even
after pollutant sources have been reduced. In open wetland systems, nutrients may also be
rapidly transported downstream, uncoupling the effects of nutrient inputs from the nutrient
source, and further complicating nutrient source control (Mitsch and Gosselink, 2000;Wetzel
2001). Recognizing relationships between nutrient input and wetland response is the first step in
10
-------
December 2006 DRAFT Chapter 1. Introduction
620 mitigating the effects of cultural eutrophication. Once relationships are established, nutrient
621 criteria can be developed to manage nutrient pollution and protect wetlands from eutrophication.
622
623
624 1.4 OVERVIEW OF THE CRITERIA DEVELOPMENT PROCESS
625
626 This section describes the five general elements of nutrient criteria development outlined in the
627 National Strategy (USEPA 1998a). A prescriptive, one-size-fits-all approach is not appropriate
628 due to regional differences that exist and the scientific community's limited technical
629 understanding of the relationship between nutrients, algal and macrophyte growth, and other
630 factors (e.g., flow, light, substrata). The approach chosen for criteria development therefore may
631 be tailored to meet the specific needs of each State or Tribe.
632
633 The USEPA is utilizing the following principal elements from the National Strategy for the
634 Development of Regional Nutrient Criteria (1998a). This document can be downloaded in PDF
635 format at the following website: http://www.epa.gov/waterscience/criteria/nutrient/strategy.html.
636
637 1. EPA will develop Ecoregional recommended nutrient criteria to account for the natural
638 variation existing across various parts of the country. Different waterbody processes and
639 responses dictate that nutrient criteria be specific to the waterbody type. No single
640 criterion is sufficient for each (or all) waterbody types; therefore, we anticipate system
§4J classification within each waterbody type for appropriate criteria derivation.
643
644 2. EPA guidance documents for nutrient criteria will provide methodologies for developing
645 nutrient criteria for primary variables by ecoregion and waterbody type.
646
647 3. Regional Nutrient Coordinators will lead State/Tribal technical and financial support
648 operations used to compile data and conduct environmental investigations. Regional
649 technical assistance groups (RTAGs) with broad participation from regional and national
650 experts on nutrients and nutrient cycling will provide technical assistance and support. A
651 team of agency specialists from USEPA Headquarters will provide additional technical
652 and financial support to the Regions, and will establish and maintain communications
653 between the Regions and Headquarters.
654
655 4. Numeric nutrient criteria, developed at the national level from existing databases and
656 additional environmental investigations, will be used by EPA to derive specific
657 recommended criterion values.
658
659 5. Nationally developed ecoregional recommended nutrient criteria may be used by States
660 and Tribes as a point of departure for the development of more refined locally and
661 regionally appropriate criteria.
662
11
-------
December 2006 DRAFT Chapter 1. Introduction
663 6. Nutrient criteria will serve as benchmarks for evaluating the relative success of any
664 nutrient management effort, whether protection or remediation is involved. EPA's
665 recommended criteria will be re-evaluated periodically to assess whether refinements or
666 other improvements are needed.
667
668 The U. S. EPA Strategy envisions a process by which State/Tribal waters are initially monitored,
669 reference conditions are established, individual waterbodies are compared to known reference
670 waterbodies, and appropriate management measures are implemented. These measurements can
671 be used to document change and monitor the progress of nutrient reduction activities.
672
673 The National Nutrient Program represents an effort and approach to criteria development that, in
674 conjunction with efforts made by State and Tribal water quality managers, will ultimately result
675 in a heightened understanding of nutrient-response relationships. As the proposed process is put
676 into use to set criteria, program success will be gauged over time through evaluation of
677 management and monitoring efforts. A more comprehensive knowledge-base pertaining to
678 nutrient, and vegetation and /or algal relationships will be expanded as new information is
679 gained and obstacles overcome, justifying potential refinements to the criteria development
680 process described here.
681
682 The overarching goal of developing nutrient criteria is to ensure the quality of our national
683 waters. Ensuring water quality may include restoration of impaired systems, conservation of high
684 quality waters, and protection of systems at high risk for future impairment. The specific goals of
685 a State or Tribal water quality program may be defined differently based on the needs of each
686 State or Tribe, but should, at a minimum, be established to protect the designated uses for the
687 waterbodies within State or Tribal lands. In addition, as numeric nutrient criteria are developed
688 for the nation's waters, States, Tribes and Territories should revisit their goals for water quality
689 and adapt their water quality standards as needed.
690
691
692 1.5 ROADMAP TO THE DOCUMENT
693
694 As set out in Figure 1.1, the process of developing numeric nutrient criteria begins with defining
695 the goals of criteria development and standards adoption. Those goals are pertinent to the
696 classification of systems, the development of a monitoring program, and the application of
697 numeric nutrient criteria to permit limits and water quality protection. These goals therefore
698 should be determined with the intent of revising and adapting them as new information is
699 obtained and the paths to achieving those goals are clarified. Defining the goals for criteria
700 development is the first step in the process. The summaries below describe each chapter in this
701 document. The document is written to provide a stepwise procedure for criteria development.
702 Some chapters contain information that is not needed by some readers; the descriptions below
703 should serve as a guide to the most relevant information for each reader.
704
12
-------
December 2006 DRAFT Chapter 1. Introduction
705 Chapter Two describes many of the functions of wetland systems and their role in the landscape
706 with respect to nutrients. This chapter is intended to familiarize the reader with some basic
707 scientific information about wetlands that will provide a better understanding of how nutrients
708 move within a wetland and the importance of wetland systems in the landscape.
709
710 Chapter Three discusses wetland classification and presents the reader with options for
711 classifying wetlands based on system characteristics. This chapter introduces the scientific
712 rationale for classifying wetlands, reviews some common classification schemes, and discusses
713 their role in establishing nutrient criteria for wetlands. The classification of these systems is
714 important to identifying their nutrient status and their condition in relation to similar wetlands.
715
716 Chapter Four provides technical guidance on designing effective sampling programs for State
717 and Tribal wetland water quality monitoring programs. Most States and Tribes should begin
718 wetland monitoring programs to collect water quality and biological data in order to develop
719 nutrient criteria protective of wetland systems. The best monitoring programs are designed to
720 assess wetland condition with statistical rigor and maximize effective use of available resources.
721 The sampling protocol selected, therefore, should be determined based on the goals of the
722 monitoring program, and the resources available.
723
724 Chapter Five gives an overview of candidate variables that could be used to establish nutrient
725 criteria for wetlands. Primary variables are expected to be most broadly useful in characterizing
726 wetland conditions with respect to nutrients and include nutrient loading rates, soil nutrient
727 concentrations, and nutrient content of wetland vegetation. Supporting variables provide
728 information useful for normalizing causal and response variables. The candidate variables
729 suggested here are not the only parameters that can be used to determine wetland nutrient
730 condition, but rather identify those variables that are thought to be most likely to identify the
731 current nutrient condition and will be most useful in determining a change in nutrient status.
732
733 A database of relevant water quality information can be an invaluable tool to States and Tribes as
734 they develop nutrient criteria. If little or no data are available for most regions or parameters, it
735 may be necessary for States and Tribes to create a database of newly gathered data. Chapter Six
736 provides the basic information on how to develop a database of nutrient information for
737 wetlands, and supplies links to ongoing database development efforts at the state and national
738 levels.
739
740 The purpose of Chapter Seven is to explore methods for analyzing data that can be used to
741 develop nutrient criteria. Proper analysis and interpretation of data determine the scientific
742 defensibility and effectiveness of the criteria. This chapter describes recommended approaches to
743 data analysis for developing numeric nutrient criteria for wetlands. Included are techniques to
744 evaluate metrics, to examine or compare distributions of nutrient exposure or response variables,
745 and to examine nutrient exposure-response relationships.
746
13
-------
December 2006 DRAFT Chapter 1. Introduction
747 Chapter Eight describes the details of setting scientifically defensible criteria in wetlands.
748 Several approaches are presented that water quality managers can use to derive numeric criteria
749 for wetland systems in their State/Tribal waters. They include: (1) the use of the reference
750 conditions concept to characterize natural or minimally impaired wetland systems with respect to
751 causal and response variables, (2) applying predictive relationships to select nutrient
752 concentrations that will protect wetland function, and (3) developing criteria from established
753 nutrient exposure-response relationships (as in the peer-reviewed published literature). This
754 chapter provides information on how to determine the appropriate numeric criterion based on the
755 data collected and analyzed.
756
757 The appendices include a glossary of terms and acronyms, and case study examples of wetland
758 nutrient enrichment and management.
14
-------
December 2006 DRAFT
Chapter 1. Introduction
759
760
Variables
Identify
Classify 1
Wetlands J
£
J
Determine^
Sampling
Etesian
Analyze
Data
Develop
Criteria
Monitor
and
761
762
rigure 13, Roue hail pro\ tiling the steps of the process to develop wetland nutrient criteria.
15
-------
December 2006- DRAFT
Chapter 2. Overview of Wetland Science
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
Chapter 2 Overview of Wetland Science
2.1 INTRODUCTION
Wetlands exist at the interface between terrestrial and aquatic environments. They serve as
sources, sinks and transformers of materials. Wetlands serve as sites for transformation of
nutrients such as nitrogen (N) and phosphorus (P). Dissolved inorganic forms of N and P are
assimilated by microorganisms and vegetation and incorporated into organic compounds. Nitrate
in surface- and ground-water is reduced to gaseous forms of N (NO, N2O, N2) by
microorganisms, a process known as denitrification, and returned to the atmosphere. Phosphorus
undergoes a variety of chemical reactions with iron (Fe), aluminum (Al), and calcium (Ca) that
depend on the pH of the soil, availability of sorption sites, redox potential and other factors.
These biogeochemical reactions are important in evaluating the nutrient condition (oligotrophic,
mesotrophic, eutrophic) of the wetland and its susceptibility to nutrient enrichment.
Atmospheric Inputs
j s Atmospheric Outputs
Surface Water Inflows
\l Ground water Inflows ,
BB
.1, Groundwater Outflows
^H
Surface Water Outflows
Figure 2.1. Schematic of nutrient transfer among potential system sources and sinks.
16
-------
December 2006- DRAFT Chapter 2. Overview of Wetland Science
782
783 Wetlands also generally are sinks for sediment, and wetlands that are connected to adjacent
784 aquatic ecosystems (e.g., rivers, estuaries) trap more sediment as compared to wetlands that lack
785 such connectivity. Wetlands also may be sources of organic carbon (C) and N to aquatic
786 ecosystems. Production of plant biomass (leaves, wood, roots) from riparian, alluvial and
787 floodplain forests and from fringe wetlands such as tidal marshes and mangroves provide organic
788 matter to support heterotrophic foodwebs of streams, rivers, estuaries and nearshore waters.
789
790
791 2.2 COMPONENTS OF WETLANDS
792
793 Wetlands are distinguished by three primary components: hydrology, soils and vegetation.
794 Wetland hydrology is the driving force that determines soil development, the assemblage of
795 plants and animals that inhabit the site, and the type and intensity of biochemical processes.
796 Wetland soils may be either organic or mineral, but share the characteristic that they are
797 saturated or flooded at least some of the time during the growing season. Wetland vegetation
798 consists of many species of algae, rooted plants that may be herbaceous and emergent, such as
799 cattail (Typha sp.) and arrowhead (Saggitaria sp.), or submergent, such as pondweeds
800 (Potamogeton sp.), or may be woody such as bald cypress (Taxodium distichuni) and tupelo
801 (Nyssa aquaticd). Depending on the duration, depth and frequency of inundation or saturation,
802 wetland plants may be either obligate (i.e., species found almost exclusively in wetlands) or
803 facultative (i.e., species found in wetlands but which also may be found in upland habitats). The
804 discussion that follows provides an overview of wetland hydrology, soils and vegetation, as well
805 as aspects of biogeochemical cycling in these systems.
806
807 HYDROLOGY
808
809 Hydrology is characterized by water source, hydroperiod (depth, duration and frequency of
810 inundation or soil saturation), and hydrodynamics (direction and velocity of water movement).
811 The hydrology of wetlands differs from that of terrestrial ecosystems in that wetlands are
812 inundated or saturated long enough during the growing season to produce soils that are at least
813 periodically deficient in oxygen. Wetlands differ from other aquatic ecosystems by their shallow
814 depth of inundation that enables rooted vegetation to become established, in contrast to deep
815 water aquatic ecosystems, where the depth and duration of inundation can be too great to support
816 emergent vegetation. Anaerobic soils promote colonization by vegetation adapted to low
817 concentrations of oxygen in the soil.
818
819 Wetlands can receive water from three sources: precipitation, surface flow and groundwater
820 (Figure 2.2). The relative proportion of these hydrologic inputs influences the plant communities
821 that develop, the type of soils that form, and the predominant biogeochemical processes.
822 Wetlands that receive mostly precipitation tend to be "closed" systems with little exchange of
823 materials with adjacent terrestrial or aquatic ecosystems. Examples of precipitation-driven
17
-------
December 2006- DRAFT
Chapter 2. Overview of Wetland Science
824
825
826
827
829
831
833
835
837
839
841
843
845
847
849
851
853
855
857
859
861
863
865
867
869
871
873
wetlands include "ombrotrophic" bogs and depressional wetlands such as cypress domes and
vernal pools. Wetlands that receive water mostly from surface flow tend to be "open" systems
with large exchanges of water
0%A100%
100%,
33%
67%
100%
and
SURFACE FLOW
Figure 2. Relationship between water source and wetland
vegetation. Modified from Brinson (1993).
877
878
879
880
881
882
883
884
885
886
887
materials between the wetland and adjacent non-wetland ecosystems. Examples include
floodplain forests and fringe wetlands such as lakeshore marshes, tidal marshes and mangroves.
Wetlands that receive primarily groundwater inputs tend to have more stable hydroperiods than
precipitation- and surface water-driven wetlands and, depending on the underlying bedrock or
parent material, high concentrations of dissolved inorganic constituents such as calcium (Ca) and
magnesium (Mg). Fen wetlands and seeps are examples of groundwater-fed wetlands.
Hydroperiod is highly variable depending on the type of wetland. Some wetlands that receive
most of their water from precipitation (e.g.,vernal pools) have very short duration hydroperiods.
Wetlands that receive most of their water from surface flooding (e.g.,floodplain swamps) often
are flooded longer and to a greater depth than precipitation-driven wetlands. Fringe wetlands
such as tidal marshes and mangroves are frequently flooded (up to twice daily) by astronomical
18
-------
December 2006- DRAFT Chapter 2. Overview of Wetland Science
889 tides but the duration of inundation is relatively short. In groundwater-fed wetlands, hydroperiod
890 is more stable and water levels are relatively constant as compared to precipitation- and surface
891 water-driven wetlands, because groundwater provides a near-constant input of water throughout
892 the year.
893
894 Hydrodynamics is especially important in the exchange of materials between wetlands and
895 adjacent terrestrial and aquatic ecosystems. In fact, the role of wetlands as sources, sinks and
896 transformers of material depends, in large part, on hydrodynamics. For example, many wetlands
897 are characterized by lateral flow of surface- or ground-water. Flow of water can be unidirectional
898 or bidirectional. An example of a wetland with unidirectional flow is a floodplain forest where
899 surface water spills over the river bank, travels through the floodplain and re-enters the river
900 channel some distance downstream. In fringe wetlands such as lakeshore marshes, tidal marshes
901 and mangroves, flow is bidirectional as wind-driven or astronomical tides transport water into,
902 then out of the wetland. These wetlands have the ability to intercept sediment and dissolved
903 inorganic and organic materials from adjacent systems as water passes through the wetland. In
904 precipitation-driven wetlands, flow may occur more in the vertical direction as rainfall percolates
905 through the (unsaturated) surface soil down to the water table. Wetlands with lateral surface flow
906 may be important in maintaining water quality of adjacent aquatic systems by trapping sediment
907 and other pollutants. Surface flow wetlands also may be an important source of organic C to
908 aquatic ecosystems as detritus, particulate C and dissolved organic C are transported out of the
909 wetland into rivers and streams down gradient or to adjacent lakes, estuaries and nearshore
910 waters.
911
912 SOILS
913
914 Wetland soils, also known as hydric soils, are defined as "soils that formed under conditions of
915 saturation, flooding or ponding long enough during the growing season to develop anaerobic
916 conditions in the upper part" (NRCS 1998). Anaerobic conditions result because the rate of
917 oxygen diffusion through water is approximately 10,000 times less than in air. Wetland soils
918 may be composed mostly of mineral constituents (sand, silt, clay) or they may contain large
919 amounts of organic matter. Because anaerobic conditions slow or inhibit decomposition of
920 organic matter, wetland soils typically contain more organic matter than terrestrial soils of the
921 same region or climatic conditions. Under conditions of near continuous inundation or
922 saturation, organic soils (histosols) may develop. Histosols are characterized by high organic
923 matter content, 20-30% (12-18% organic C depending on clay content) with a thickness of at
924 least 40 cm (USDA 1999). Because of their high organic matter content, Histosols possess
925 physical and chemical properties that are much different from mineral wetland soils. For
926 example, organic soils generally have lower bulk densities, higher porosity, greater water
927 holding capacity, lower nutrient availability, and greater cation exchange capacity than many
928 mineral soils.
929
19
-------
December 2006- DRAFT Chapter 2. Overview of Wetland Science
930 Mineral wetland soils, in addition to containing greater amounts of sand, silt and clay than
931 histosols, are distinguished by changes in soil color that occur when elements such as Fe and
932 manganese (Mn) are reduced by microorganisms under anaerobic conditions. Reduction of Fe
933 leads to the development of grey or "gleyed" soil color as oxidized forms of Fe (ferric Fe, Fe3+)
934 are converted to reduced forms (ferrous Fe, Fe+2). In sandy soils, development of a dark-colored
935 organic-rich surface layer is used to distinguish hydric soil from non-hydric (terrestrial) soil. An
936 organic-rich surface layer, indicative of periodic inundation or saturation, is not sufficiently thick
937 (<40 cm) to qualify as a histosol which forms under near-continuous inundation.
938
939 Wetland soils serve as sites for many biogeochemical transformations. They also provide long
940 and short term storage of nutrients for wetland plants. Wetland soils are typically anaerobic
941 within a few millimeters of the soil-water interface. Water column oxygen concentrations are
942 often depressed due to the slow rate of oxygen diffusion through water. However, even when
943 water column oxygen concentrations are supported by advective currents, high rates of oxygen
944 consumption lead to the formation of a very thin oxidized layer at the soil-water interface.
945 Similar oxidized layers can also be found surrounding roots of wetland plants. Many wetland
946 plants are known to transport oxygen into the root zone, thus creating aerobic zones in
947 predominantly anaerobic soil. The presence of these aerobic (oxidizing) zones within the
948 reducing environment in saturated soils allows for the occurrence of oxidative and reductive
949 transformations to occur in close proximity to each other. For example, ammonia is oxidized to
950 nitrate within the aerobic zone surrounding plant roots in a process called nitrification. Nitrate
951 then readily diffuses into adjacent anaerobic soil, where it is reduced to molecular nitrogen via
952 denitrification or may be reduced to ammonium in certain conditions through dissimilatory
953 nitrate reduction (Mitsch and Gosselink 2000; Ruckauf et al., 2004; Reddy and Delaune, 2005).
954 The anaerobic environment hosts the transformations of N, P, sulfur (S), Fe, Mn, and C. Most of
955 these transformations are microbially mediated. The oxidized soil surface layer also is important
956 to the transport and translocation of transformed constituents, providing a barrier to translocation
957 of some reduced constituents. These transformations will be discussed in more detail below in
958 Biogeochemical Cycling.
959
960 VEGETATION
961
962 Wetland plants consist of macrophytes and microphytes. Macrophytes include free-floating,
963 submersed, floating-leaved and rooted emergent plants. Microphytes are algae that may be free
964 floating or attached to macrophyte stems and other surfaces. Plants require oxygen to meet
965 respiration demands for growth, metabolism and reproduction. In macrophytes, much (about
966 50%) of the respiration occurs below ground in the roots. Wetland macrophytes, however, live in
967 periodically to continuously-inundated and saturated soils and, so, use specialized adaptations to
968 grow in anaerobic soils. Adaptations consist of morphological/anatomical adaptations that result
969 in anoxia avoidance and metabolic adaptations that result in true tolerance to anoxia.
970 Morphological/anatomic adaptations include shallow roots systems, aerenchyma, buttressed
971 trunks, pneumatophores (e.g., cypress "knees") and lenticles on the stem. These adaptations
20
-------
December 2006- DRAFT Chapter 2. Overview of Wetland Science
972 facilitate oxygen transport from the shoots to the roots where most respiration occurs. Many
973 wetland plants also possess metabolic adaptations, such as anaerobic pathways of respiration,
974 that produce non-toxic metabolites such as malate to mitigate the adverse effects of oxygen
975 deprivation, instead of toxic compounds like ethanol (Mendelssohn and Burdick 1988).
976
977 Species best adapted to anaerobic conditions are typically found in areas inundated for long
978 periods, whereas species less tolerant of anaerobic conditions are found in areas where
979 hydroperiod is shorter. For example, in southern forested wetlands, areas such as abandoned
980 river channels (oxbows) are dominated by obligate species such as bald cypress (Taxodium
981 distichuni) and tupelo gum (Nyssa aquaticd) (Wharton et al. 1982). Areas inundated less
982 frequently are dominated by hardwoods such as black gum (Nyssa sylvatica), green ash
983 (Fraxinuspennsylvanicus) and red maple (Acer rubrum) and the highest, driest wetland areas are
984 dominated by facultative species such as sweet gum (Liquidambar styraciflua) and sycamore
985 (Platanus occidentalis) (Wharton et al. 1982). Herbaceous-dominated wetlands also exhibit
986 patterns of zonation controlled by hydroperiod (Mitsch and Gosselink 2000).
987
988 In estuarine wetlands such as salt- and brackish-water marshes and mangroves, salinity and
989 sulfides also adversely affect growth and reproduction of vegetation. Inundation with seawater
990 brings dissolved salts (NaCl) and sulfate. Salt creates an osmotic imbalance in vegetation,
991 leading to dessication of plant tissues. However, many plant species that live in estuarine
992 wetlands possess adaptations to deal with salinity (Whipple et al., 1981; Zheng et al. 2004).
993 These adaptations include salt exclusion at the root surface, salt secreting glands on leaves,
994 schlerophyllous (thick, waxy) leaves, low transpiration rates and other adaptations to reduce
995 uptake of water and associated salt. Sulfate carried in by the tides undergoes sulfate reduction in
996 anaerobic soils to produce hydrogen sulfide (ftzS) that, at high concentrations, is toxic to
997 vegetation. At sub-lethal concentrations, H2S inhibits nutrient uptake and impairs plant growth.
998
999 SOURCES OF NUTRIENTS
1000
1001 Point Sources
1002 Point source discharges of nutrients to wetlands may come from municipal or industrial
1003 discharges, including stormwater runoff from municipalities or industries, or in some cases from
1004 large animal feeding operations. Nutrients from point source discharges may be controlled
1005 through the National Pollutant Discharge Elimination System (NPDES) permits, most of which
1006 are administered by states authorized to issue such permits. In general, point source discharges
1007 that are not stormwater related are fairly constant with respect to loadings.
1008
1009 Nonpoint Sources
1010 Nonpoint sources of nutrients are commonly discontinuous and can be linked to seasonal
1011 agricultural activity or other irregularly occurring events such as silviculture, non-regulated
1012 construction, and storm events. Nonpoint nutrient pollution from agriculture is most commonly
1013 associated with row crop agriculture, and livestock production that tend to be highly associated
21
-------
December 2006- DRAFT Chapter 2. Overview of Wetland Science
1014 with rain events and seasonal land use activities. Nonpoint nutrient pollution from urban and
1015 suburban areas is most often associated with climatological events (rain, snow, and snowmelt)
1016 when pollutants are most likely to be transported to aquatic resources.
1017
1018 Runoff from agricultural and urban is generally thought to be the largest source of nonpoint
1019 source pollution; however growing evidence suggests that atmospheric deposition may have a
1020 significant influence on nutrient enrichment, particularly from nitrogen (Jaworski et al. 1997).
1021 Gases released through fossil fuel combustion and agricultural practices are two major sources of
1022 atmospheric N that may be deposited in waterbodies (Carpenter et al. 1998). Nitrogen and
1023 nitrogen compounds formed in the atmosphere return to the earth as acid rain or snow, gas, or
1024 dry particles. Atmospheric deposition, like other forms of pollution, may be determined at
1025 different scales of resolution. More information on national atmospheric deposition can be found
1026 at: http://www.arl.noaa.gov/research/programs/airmon.html; http://nadp.sws.uiuc.edu/. These
1027 national maps may provide the user with information about regional areas where atmospheric
1028 deposition, particularly of nitrogen, may be of concern. However, these maps are generally low
1029 resolution when considered at the local and site-specific scale and may not reflect areas of high
1030 local atmospheric deposition, such as local areas in a downwind plume from an animal feedlot
1031 operation.
1032
1033 Other nonpoint sources of nutrient pollution may include certain silviculture and mining
1034 operations; these activities generally constitute a smaller fraction of the national problem, but
1035 may be locally significant nutrient sources. Control of nonpoint source pollutants focuses on land
1036 management activities and regulation of pollutants released to the atmosphere (Carpenter et al.
1037 1998).
1038
1039
1040 2.3 WETLAND NUTRIENT COMPONENTS
1041
1042 NUTRIENT BUDGETS
1043
1044 Wetland nutrient inputs mirror wetland hydrologic inputs (e.g., precipitation, surface water, and
1045 ground water), with additional loading associated with atmospheric dry deposition and
1046 nitrification (Figures 2.5 and 2.6). Total atmospheric deposition (wet and dry) may be the
1047 dominant input for precipitation-dominated wetlands, while surface- or ground-water inputs may
1048 dominate other wetland systems.
1049
1050 The total annual nutrient load (mg-nutrients/yr) into a wetland is the sum of the dissolved and
1051 particulate loads. The dissolved load (mg-nutrients/s) can be estimated by multiplying the
1052 instantaneous inflow (L/s) by the nutrient concentration (mg-nutrients/L). EPA recommends
1053 calculating the annual load by the summation of this function over the year - greater loads may
1054 found during periods of increased flow so EPA recommends monitoring during these intervals.
1055 Where continuous data are unavailable, average flows and concentrations may be used if a bias
22
-------
December 2006- DRAFT Chapter 2. Overview of Wetland Science
1056 factor (Cohn et al., 1989) is included to account for unmeasured loads during high flows.
1057 Particulate loads (kg-sediments/yr) can be estimated using the product of suspended and bedload
1058 inputs (kg-sediments/yr) and the mass concentrations (mg-nutrients/kg-sediment).
1059
1060 Surface-water nutrient inputs are associated with flows from influent streams, as well as diffuse
1061 sources from overland flow through the littoral zone. Ground-water inputs can also be
1062 concentrated at points (e.g., springs), or diffuse (such as seeps). The influence of allochthonous
1063 sources is likely to be greatest in those zones closest to the source.
1064
1065 Because wetlands generally tend to be low-velocity, depositional environments, they often
1066 sequester sediments and their associated nutrients. These sediment inputs generally accumulate
1067 at or near the point of entry into the wetland, forming deltas or levees near tributaries, or along
1068 the shoreline for littoral inputs. Coarser fractions (e.g., gravels and sands) tend to settle first,
1069 with the finer fractions (silts, clays, and organic matter) tend to settle further from the inlet point.
1070 Particulate input from ground-water sources can usually be neglected, while particulate inputs
1071 from atmospheric sources may be important if local or regional sources are present.
1072
1073 Wetland nutrient outputs again mirror hydrologic outputs (e.g., surface- and ground-water), and
1074 loads are again estimated as the product of the flow and the concentration of nutrients in the
1075 flow. While evaporation losses from wetlands may be significant, there are no nutrient losses
1076 associated with this loss. Instead, loss of nutrients to the atmosphere may occur as a result of
1077 ammonia volatilization, N2O losses from nitrification, as well as losses from incomplete
1078 denitrification. Because sediment outputs from wetlands may be minor, nutrient exports by this
1079 mechanism may not be important.
1080
1081 Nutrient accumulation in wetlands occurs when nutrient inputs exceed outputs. Net nutrient
1082 loads can be estimated as the difference between these inputs and outputs. It is important,
1083 therefore, to have some estimate of net accumulation by taking the difference between upstream
1084 and downstream loads. Sampling ground-water nutrient concentrations in wells located upstream
1085 and downstream of the wetland can provide some sense of net nutrient sequestration, while
1086 sampling wetland nutrient inflows and outflows is needed for determining the additional
1087 sequestration for this pathway.
1088
1089 BlOGEOCHEMICAL CYCLING
1090
1091 Biogeochemical cycling of nutrients in wetlands is governed by physical, chemical and
1092 biological processes in the soil and water column. Biogeochemical cycling of nutrients is not
1093 unique to wetlands, but the aerobic and anaerobic interface generally found in saturated soils of
1094 wetlands creates unique conditions that allow both aerobic and anaerobic processes to operate
1095 simultaneously. The hydrology and geomorphology of wetlands (Johnston et al. 2001) influences
1096 biogeochemical processes and constituent transport and transformation within the systems (e.g.,
1097 water-sediment exchange, plant uptake, and export of organic matter). Interrelationships among
23
-------
December 2006- DRAFT Chapter 2. Overview of Wetland Science
1098 hydrology, biogeochemistry, and the response of wetland biota vary among wetland types
1099 (Mitsch and Gosselink, 2000; Reddy and Delaune, 2005).
1100
1101 Biogeochemical processes in the soil and water column are key drivers of several ecosystem
1102 functions associated with wetland values (e.g., water quality improvement through
1103 denitrification, long-term nutrient storage in the organic matter) (Figure 2.3). The hub for
1104 biogeochemistry is organic matter and its cycling in the soil and water column. Nutrients such as
1105 N, P, and S are primary components of soil organic matter, and cycling of these nutrients is
1106 always coupled to C cycling. Many processes occur within the carbon, nitrogen, phosphorus and
1107 sulfur (C, N, P, or S) cycles; microbial communities mediate the rate and extent of these
1108 reactions in soil and the water column.
1109
1110 Aerobic-anaerobic interfaces are more common in wetlands than in upland landscapes and may
1111 occur at the soil-water interface, in the root zones of aquatic macrophytes, and at surfaces of
1112 detrital tissue and benthic periphyton mats. The juxtaposition of aerobic and anaerobic zones in
1113 wetlands supports a wide range of microbial populations and associated metabolic activities,
1114 with oxygen reduction occurring in the aerobic interface of the substrate, and reduction of
1115 alternate electron acceptors occurring in the anaerobic zone (D'Angelo and Reddy, 1994a or b).
1116 Under continuously saturated soil conditions, vertical layering of different metabolic activities
1117 can be present, with oxygen reduction occurring at and just below the soil-floodwater interface.
1118 Substantial aerobic decomposition of plant detritus occurs in the water column; however, the
1119 supply of oxygen may be insufficient to meet demands and drive certain microbial groups to
1120 utilize alternate electron acceptors (e.g., nitrate, oxidized forms of iron (Fe) and manganese
1121 (Mn), sulfate and bicarbonate (HCO3)).
1122
24
-------
December 2006- DRAFT
Chapter 2. Overview of Wetland Science
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
Submerged
macrophyte
Emergent
macrophyte
Periphyton
"Water
Bioavailable
nutrients
Figure 2.3 Schematic showing basic nutrient cycles in soil-water column of a wetland.
Soil drainage adds oxygen to the soil, while other inorganic electron acceptors may be added
through hydraulic loading to the system. Draining wetland soil accelerates organic matter
decomposition due to the introduction of oxygen deeper into the profile. In many wetlands, the
influence of NOs, and oxidized forms of Mn and Fe on organic matter decomposition is minimal.
This is because the concentrations of these electron acceptors are usually low as a result of the
fact that they have greater reduction potential than other alternate electron acceptors, so they
generally are depleted rapidly from systems. Long-term sustainable microbial activity is then
supported by electron acceptors of lower reduction potentials (sulfate and HCOs).
Methanogenesis is often viewed as the terminal step in anaerobic decomposition in freshwater
wetlands, whereas sulfate reduction is viewed as the dominant process in coastal wetlands.
However, both processes can function simultaneously in the same ecosystem and compete for
available substrates (Capone and Kiene 1988).
A simple way to characterize wetlands for aerobic and anaerobic zones is to determine the
oxidation-reduction potential or redox potential (Eh) of the soil-water column. Redox potential is
expressed in units of millivolts (mV) and is measured using a voltmeter coupled to a platinum
electrode and a reference electrode. Typically, wetland soils with Eh values >300 mV are
25
-------
December 2006- DRAFT
Chapter 2. Overview of Wetland Science
1143
1144
1145
considered aerobic and typical of drained soil conditions, while soils with Eh values <300 mV
are considered anaerobic and are devoid of molecular oxygen (Figure 2.4).
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
-3
Ana
DIL [
prnhic 4
H'9h|y Reduced
Reduced
1 1 1 1
Moderately
Reduced
fc Aorohic
Oxidized
I I I I I I I
I
00 -100 0 100 300 500 700
Oxidation-Reduction Potential (mV)
Figure 2.4. Range of redox potentials in wetland soils (Reddy and Delaune 2007).
Wetlands, as low-lying areas in the landscape, receive inputs from all hydrologically connected
uplands. Many wetlands are open systems receiving inputs of carbon (C) and nutrients from
upstream portions of the watershed that can include agricultural and urban areas.
Prolonged nutrient loading to wetlands can result in distinct gradients in water and soil. Mass
loading and hydraulic retention time determine the degree and extent of nutrient enrichment.
Continual nutrient loading to an oligotrophic wetland can result in a zone of high nutrient
availability near the input, and low nutrient availability and possibly nutrient limiting conditions
further from the input point. This enrichment effect can be seen in many freshwater wetlands,
most notably in the sub-tropical Everglades where light is abundant and temperatures are high
(Davis, 1991; Reddy et al., 1993; Craft and Richardson, 1993 a, b; DeBusk et al., 1994) and in
some estuarine marshes (Morris and Bradley 1999). Between these two extremes, there can exist
26
-------
December 2006- DRAFT
Chapter 2. Overview of Wetland Science
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
a gradient in quality and quantity of organic matter, nutrient accumulation, microbial and
macrobiotic communities, composition, and biogeochemical cycles.
Compared to terrestrial ecosystems, most wetlands show an accumulation of organic matter, and
therefore wetlands function as global sinks for carbon. Accumulation of organic C in wetlands is
primarily a result of the balance of C fixation through photosynthesis and losses through
decomposition. Rates of photosynthesis in wetlands are typically higher than in other ecosystems,
and rates of decomposition are typically lower due to anaerobic conditions, hence organic matter
tends to accumulate. In addition to maintaining proper functioning of wetlands, organic matter
storage also plays an important role in regulating other ecosystems and the biosphere. For example,
organic matter contains substantial quantities of N, P, and S, therefore accumulation of organic
matter in wetlands decreases transport of these nutrients to downstream aquatic systems.
NITROGEN (N):
Nitrogen enters wetlands in organic and inorganic forms, with the relative proportion of each
depending on the input source. Organic forms are present in dissolved and particulate fractions,
while inorganic N (NFL^N, NOs-N and NO2-N) is present in dissolved fractions (Fig. 2.5) or
bound to suspended sediments (NH4-N). Particulate fractions are removed through settling and
burial, while the removal of dissolved forms is regulated by various biogeochemical reactions
functioning in the soil and water column. Relative rates of these processes are affected by
physico-chemical and biological characteristics of plants, algae and microorganisms.
Atmospheric
Deposition
Plant biomass
N2 N2O
^Inflow
Outflow
NH,
Litterfall
Volatilization
Nitrogen
Fixation
Nitrification
NO3 < NH4
Mineralization
WaterColumn
Soil - AEROBIC
[NH4+]
Denitrification
Microbial
Organic N •**• Biomass N Adsorbed NH4+
Soil - ANAEROBIC
— N2, N20 (g)
Figure 2.5. Schematic of the nitrogen cycle in wetlands.
27
-------
December 2006- DRAFT Chapter 2. Overview of Wetland Science
1188 Nitrogen reactions in wetlands effectively process inorganic N through nitrification and
1189 denitrification, ammonia volatilization and plant uptake. These processes aid in lowering levels
1190 of inorganic N in the water column. A significant portion of dissolved organic N assimilated by
1191 plants is returned to the water column during breakdown of detrital tissue or soil organic matter,
1192 and the majority of this dissolved organic N is resistant to decomposition. Under these
1193 conditions, water leaving wetlands may contain elevated levels of N in organic form. Exchange
1194 of dissolved nitrogen species between soil and water column support several nitrogen reactions.
1195 For example, nitrification in the aerobic soil layer is supported by ammonium flux from the
1196 anaerobic soil layer. Similarly, denitrification in the anaerobic soil layer is supported by nitrate
1197 flux from the aerobic soil layer and water column. Relative rates of these reactions will,
1198 however, depend on the environmental conditions present in the soil and water column (Reddy
1199 and Delaune 2007).
1200
1201 PHOSPHORUS (P):
1202
1203 Phosphorus retention by wetlands is regulated by physical (sedimentation and entrainment),
1204 chemical (precipitation and flocculation), and biological mechanisms (uptake and release by
1205 vegetation, periphyton and microorganisms). Phosphorus in the influent water is found in
1206 soluble and particulate fractions, with both fractions containing a certain proportion of inorganic
1207 and organic forms. Relative proportions of these pools depend on the input source. For example,
1208 municipal wastewater may contain a large proportion (>75%) as inorganic P in soluble forms, as
1209 compared to effluents from agricultural watersheds where a greater percentage of P loading may
1210 be in the parti culate fraction.
1211
1212 Phosphorus forms that enter a wetland are grouped into: (i) dissolved inorganic P (DIP), (ii)
1213 dissolved organic P (DOP), (iii) particulate inorganic P (PIP), and (iv) particulate organic P
1214 (POP) (Figure 2.6). The particulate and soluble organic fractions may be further separated into
1215 labile and refractory components. Dissolved inorganic P is generally bioavailable, whereas
1216 organic and particulate P forms generally must be transformed into inorganic forms before
1217 becoming bioavailable. Both biotic and abiotic mechanisms regulate relative pool sizes and
1218 transformations of P compounds within the water column and soil. Alterations in these fractions
1219 can occur during flow through wetlands and depend on the physical, chemical, and biological
1220 characteristics of the systems. Thus, both biotic and abiotic processes should be considered when
1221 evaluating P retention capacities of wetlands. Biotic processes include; assimilation by
1222 vegetation, plankton, periphyton and microorganisms. Abiotic processes include sedimentation,
1223 adsorption by soils, precipitation, and exchange processes between soil and the overlying water
1224 column (Reddy et al. 1999; 2005; Reddy and Delaune, 2007). The processes affecting
1225 phosphorus exchange at the soil/sediment water interface include: (i) diffusion and advection due
1226 to wind-driven currents, (ii) diffusion and advection due to flow and bioturbation, (iii) processes
1227 within the water column (mineralization, sorption by particulate matter, and biotic uptake and
1228 release), (iv) diagenetic processes (mineralization, sorption and precipitation dissolution) in
28
-------
December 2006- DRAFT
Chapter 2. Overview of Wetland Science
1229
1230
1231
1232
bottom sediments, (v) redox conditions (C>2 content) at the soil/sediment -water interface, and
(vi) phosphorus flux from water column to soil mediated by evapotranspiration by vegetation.
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
Atmospheric
Deposition
Plant biomass P
Figure 2.6. Phosphorous cycle in wetlands.
The key biogeochemical services provided by wetlands include nutrient transformation and
removal by decreasing concentrations of nutrients and other contaminants and sequestration of
carbon and nutrients into stable pools (Kadlec and Knight 1996). The biogeochemical processes
regulating water quality improvement are well established, and are made use of in treatment
wetlands. Increased nutrient loading to oligotrophic wetlands results in increased primary
productivity and nutrient enrichment. This resulting eutrophication can have both positive and
negative impacts to the environment. Higher rates of primary productivity increase rates of
organic matter accumulation, thus increasing carbon sequestration. However eutrophication
may lead to increased periodic and episodic export of DIP (Kadlec and Knight 1996; Reddy et
al. 1995; 2005; Reddy and Delaune 2007)).
29
-------
December 2006- DRAFT Chapter 3. Classification of Wetlands
1247 Chapter 3 Classification of Wetlands
1248
1249
1250 3.1 INTRODUCTION
1251
1252 Developing individual, site-specific nutrient criteria is not practical for every wetland. Instead,
1253 criteria for groups of similar wetlands in a region are needed. To this end, a means of grouping
1254 or classifying wetlands is needed. This chapter introduces the scientific rationale for classifying
1255 wetlands, reviews some common classification schemes, and discusses their implications for
1256 establishing nutrient criteria for wetlands. Use of a common scheme across state boundaries
1257 should facilitate collaborative efforts in describing reference condition for biota or water quality
1258 and in developing assessment methods, indices of biotic integrity (IBI) (USEPA 1993b,
1259 http://www.epa.gov/emap/remap/index.html), nutrient-response relationships, and nutrient
1260 criteria for wetlands. This chapter describes a series of national classification systems that could
1261 be used to provide a common framework for development of nutrient criteria for wetlands, and
1262 suggest ways in which these classification schemes could be combined in a hierarchical fashion.
1263 Many existing classification schemes may be relevant and should be considered for use or
1264 modification even if they weren't originally derived for wetland nutrient criteria because 1) they
1265 incorporate key factors which control nutrient inputs and cycling; 2) they already have been
1266 mapped; and 3) they have been incorporated into sampling, assessment, and management
1267 strategies for wetland biology or for other surface water types, thus facilitating integration of
1268 monitoring strategies. Adoption of any classification scheme should be an iterative process,
1269 whereby initial results of biological or water quality sampling are used to test for actual
1270 differences in reference condition for nutrients or nutrient-response relationships among
1271 proposed wetland classes. Wetland classes that behave similarly can be combined, and apparent
1272 outliers in distributions of nutrient concentrations from reference sites or in nutrient-response
1273 relationships can be examined for additional sources of variability that need to be considered. In
1274 addition, new classification schemes can be derived empirically through many multivariate
1275 statistical methods designed to determine factors that can discriminate among wetlands based on
1276 nutrient levels or nutrient-response relationships.
1277
1278 The overall goal of classification is to reduce variability within classes due to differences in
1279 natural condition related to factors such as geology, hydrology, and climate. This will minimize
1280 the number of classes for which reference conditions should be defined. For example, we would
1281 expect different conditions for water quality or biological community composition for wetland
1282 classes in organic soils (histosols) compared to wetlands in mineral soils. In assessing impacts to
1283 wetlands, comparing a wetland from within the same class would increase the precision of
1284 assessments, enable more sensitive detection of change, and reduce errors in characterizing the
1285 status of wetland condition.
1286
1287
30
-------
December 2006- DRAFT Chapter 3. Classification of Wetlands
1288 REFERENCE CONCEPT
1289
1290 Reference conditions "describe the characteristics of waterbody segments least impaired by
1291 human activities and are used to define attainable biological or habitat conditions" (USEPA
1292 1990, Stoddard et al. 2006). At least two general approaches have been defined to establish
1293 reference condition - the site-specific approach and the regional approach (U.S. EPA 1990b,
1294 http://www.epa.gov/bioindicators/). The current approach to developing water quality criteria for
1295 nutrients also emphasizes identification of expected ranges of nutrients by waterbody type and
1296 ecoregion for least-impaired reference conditions (U.S. EPA 1998,
1297 http ://www. epa. gov/waterscience/standards/nutrient.html).
1298
1299 Although different concepts of reference condition have been used in other programs (e.g., for
1300 evaluation of wetland mitigation projects (Smith et al. 1995;
1301 http://el.erdc.usace.army.mil/wetlands/pdfs/wrpde9.pdf)), for the purposes of this document, the
1302 term reference condition refers to wetlands that are minimally or least impacted by human
1303 activities. Most, if not all, wetlands in the U.S. are affected to some extent by human activities
1304 such as acid precipitation, global climate change, or other atmospheric deposition of nitrogen
1305 and mercury, and changes in historic fire regime. "Minimally impacted" is therefore
1306 operationally defined by choosing sites with fewer stressors or fewer overall impacts as
1307 described by indicators of stressors, such as land-use or human activities within the watershed or
1308 buffer area surrounding a wetland and source inputs. Identifying reference wetlands in areas of
1309 high local or regional atmospheric deposition of nitrogen should also be carefully considered
1310 because indicators such as local land use activities may not be sufficient to indicate nutrient
1311 enrichment from dry or wet air deposition.
1312
1313
1314 3.2 EXISTING WETLAND CLASSIFICATION SCHEMES
1315
1316 There are two different approaches for classification of aquatic resources, one that is
1317 geographically-based, and one that is independent of geography, but relies on environmental
1318 characteristics that determine aquatic ecosystem status and vulnerability at the region-,
1319 watershed-, or ecosystem-scale (Detenbeck et al. 2000). Ecoregions (including "nutrient
1320 ecoregions") and Ecological Units represent geographically-based classification schemes that
1321 have been developed and applied nation-wide (Omernik 1987, Keys et al. 1995). The goal of
1322 geographically-based classification schemes is to reduce variability in reference condition based
1323 on spatial co-variance in climate and geology, along with topography, vegetation, hydrology, and
1324 soils. Geographically-independent or environmentally-based classification schemes include those
1325 derived using watershed characteristics such as land-use and/or land-cover (Detenbeck et al.
1326 2000), hydro geomorphology (Brinson 1993), vegetation type (Grossman et al. 1998), or some
1327 combination of these (Cowardin et al. 1979). Both geographically-based and environmentally-
1328 based schemes have been developed for wetland classification. These approaches can be applied
1329 individually or combined within a hierarchical framework (Detenbeck et al. 2000).
31
-------
December 2006- DRAFT Chapter 3. Classification of Wetlands
1330
1331 GEOGRAPHICALLY-BASED CLASSIFICATION SCHEMES
1332
1333 Regional classification systems were first developed specifically for the United States by land
1334 management agencies. The US Department of Agriculture (USDA) has described an hierarchical
1335 system of Land Resource Regions and Major Land Resource Areas based mainly on soil
1336 characteristics for agricultural management (USDA SCS 1981). Ecoregions were then refined for
1337 USDA and the US Forest Service based on an hierarchical system in which each of several
1338 environmental variables such as climate, landform, and potential natural vegetation were applied
1339 to define different levels of classification (Bailey 1976). Subsequently, Omernik and colleagues
1340 developed an hierarchical nationwide ecoregion system to classify streams, using environmental
1341 features they expected to influence aquatic resources as opposed to terrestrial resources (Hughes
1342 and Omernik 1981, Omernik et al. 1982). The latter was based on an overlay of "component
1343 maps" for land use, potential natural vegetation, land-surface form, and soils along with a
1344 subjective evaluation of the spatial congruence of these factors as compared to the hierarchical
1345 approach used by Bailey, which relied only on natural features (not land-use). Omernik has
1346 produced a national map of 84 ecoregions defined at a scale of 1:7,500,000 (Figure 3.1; Omernik
1347 1987, http://water.usgs.gov/GIS/metadata/usgswrd/XML/ecoregion.xml). More detailed,
1348 regional maps have been prepared at a scale of 1:2,500,000 in which the most "typical" areas
1349 within each ecoregion are defined. Cowardin et al. (1979) have suggested an amendment to
1350 Bailey's ecoregions to include coastal and estuarine waters (Figure 3.2). In practice, Omernik's
1351 scheme has been more widely used for geographic classification of aquatic resources such as
1352 streams, but few examples to verify the appropriateness of this grouping to wetland nutrients are
1353 available.
32
-------
December 2006- DRAFT
Chapter 3. Classification of Wetlands
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
Figure 3.1 Map of Omernik aquatic ecoregions.
5. WEST
INDIAN
BOUNDARIESOF
10 MARINE AMD ESTUARINE
PROVINCES
Figure 3.2a. Map of Bailey ecoregions with coastal and estuarine provinces
(Cowardin et al., 1979).
33
-------
December 2006- DRAFT
Chapter 3. Classification of Wetlands
1377
Figure 3.2b. Legend
"Domains, Divisions, Provinces, and Sections used on Bailey's (1976) map and described in detail in Bailey
(1978). Highland ecoregions are designated M mountain, P plateau, and A altiplano.
1000 Polar
1200 Tundra
1210 Arctic Tundra
1220 Bering Tundra
M1210 Brooks Range
1300 Subarctic
1310 Yukon Parkland
1320 Yukon Forest
M1310 Alaska Range
2000 Humid Temperate
2100 Warm Continental
2110 Laurentian Mixed Forest
2111 Spruce-Fir Forest
2112 Northern Hardwoods-Fir Forest
2113 Northern Hardwoods Forest
2114 Northern Hardwoods-Spruce Forest
M2110 Columbia Forest
M2111 Douglas-fir Forest
M2112 Cedar-Hemlock-Douglas-fir Forest
2200 Hot Continental
2210 Eastern Deciduous Forest
2211 Mixed Mesophytic Forest
2212 Beech-Maple Forest
2213 Maple-Basswoocl Forest + Oak Savanna
2214 Appalachian Oak Forest
2215 Oak-Hickory Forest
2300 Subtropical
2310 Outer Coastal Plain Forest
2311 Beech-Sweetgum-Magnolia-Pine-Oak
2312 Southern Floodplain Forest
2320 Southeastern Mixed Forest
2400 Marine
2410 Willamette-Puget Forest
M2410 Pacific Forest (in conterminous U.S.)
M2411 Sitka Spruce-Cedar-Hemlock Forest
M2412 Redwood Forest
M2413 Cedar-Hemlock-Douglas-fir Forest
M2414 California Mixed Evergreen Forest
M2415 Silver fir-Douglas-fir Forest
M2410 Pacific Forest (in Alaska)
2500 Prairie
2510 Prairie Parkland
2511 Oak-Hickory-Bluestem Parkland
2512 Oak + Bluestem Parkland
1378
2520 Prairie Brushland
2521 Mesquite-Buffalo Grass
2522 Juniper-Oak-Mesquite
2523 Mesquite-Acacia
2530 Tall-Grass Prairie
2531 Bluestem Prairie
2532Whestgrass-Bluestem-Needlegrass
2533 Bluestem-Gamma Prairie
2600 Mediterranean (Dry-summer Subtropical)
2610 California Grassland
M2610Sierran Forest
M2620 California Chaparral
3000 Dry 3100 Steppe
3110 Great Plains-Shortgrass Prairie
3111 Gramma-Needlegrass-Wheatgrass
3112 Wheatgrass-Needlegrass
3113Grama-BuffaloGrass
M3110 Rocky Mountain Forest
M3111 Grand-fir-Douglas-fir Forest
M3112 Douglas-fir Forest
M3113 Ponderosa Pine-Douglas-fir Forest
3120 Palouse Grassland
M3120 Upper Gila Mountains Forest
3130 Intermountain Sagebrush
3131 Sagebrush-Wheatgrass
3132 LahontanSaltbush-Greasewood
3133 Great Basin Sagebrush
3134 Bonneville Saltbush-Greasewood
3135 Ponderosa Shrub Forest
P3130 Colorado Plateau
P3131 Juniper-Pinyon Woodland + Sagebrush Saltbush Mosaic
P3132 Grama-Galleta Steppe + Juniper-Pinyon Woodland Mosaic
3140 Mexican Highland Shrub Steppe
A3140 Wyoming Basin
A3141 Wheatgrass- Need legrass-Sagebrush
A3142 Sagebrush-Wheatgrass
3200 Desert 3210 Chihuahuan Desert
3211 Grama-Tobosa
3212Tarbush-CreosoteBush
3220 American Desert
3221 Creosote Bush
3222 Creosote Bush-Bur Sage
4000 Humid Tropical
4100 Savanna
4110 Everglades
4200 Rainforest
M4210 Hawaiian Islands
34
-------
December 2006- DRAFT
Chapter 3. Classification of Wetlands
1379
1380
1381
1382
1383
1384
a. Domain
e. Province
b. Division
231 Southeastern
Mixed Fccest
d. Section
231A. Southern
Appalachian Piedmon
Figure 3.3 Examples of first four hierarchical levels of Ecological Units:
domain, division, province, and section, from USEPA Environmental
Atlas.
35
-------
December 2006- DRAFT
Chapter 3. Classification of Wetlands
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
Figure 3.4. Dominant water sources to wetlands, from Brinson 1993.
Finally, an attempt has been made to integrate approaches across Federal agencies to produce
regional boundaries termed Ecological Units (Keys et al. 1995). Information has been combined
on climate, landform, geomorphology, geology, soils, hydrology, and potential vegetation to
produce a nested series of boundaries for the eastern U.S. Different combinations of
environmental parameters are emphasized at each hierarchical level of classification. This
scheme was developed to explain variation in both terrestrial and aquatic systems, and is
consistent with a more comprehensive strategy to classify lotic systems down to the level of
stream reaches (Maxwell et al. 1995). The mapped system for the eastern U.S. includes
classification at the following levels:
domain (n=2) > divisions (n=5) > provinces (n=14) > sections (n=78) > subsections (n=xxx),
where Sections are roughly half the size of Omernik ecoregions (Figure 3.3). For lotic systems,
additional spatial detail can be added by defining watersheds (at the level of land type
associations), subwatersheds (at the level of land types), valley segments, stream reaches, and
finally channel units (Maxwell et al. 1995). In reality, not all watersheds nest neatly within
subsections, and may cross-subsection boundaries.
36
-------
December 2006- DRAFT
Chapter 3. Classification of Wetlands
VERTICAL
f-LUCTUATIONS
UNIDIRECTIONAL
FLOW
BIDIRECTIONAL
FLOW
1408
1409 Figure 3.5. Dominant hydrodynamic regimes for wetlands based on flow pattern (Brinson 1993).
1410
LAND SURFACE
SEEPAGE FACE
a. Seepage tace where grourxtaaitir do* Inlersecls. I tie land surlaca
1411
1412
1413
1414
1415
1416
LINES OF EQUAL o a o
HYDRAULIC HEAD
SEEP ACE AT HASE
S S
b. Seepage in the Idwere ooe portion of the bre
Figure 3.6. Interaction with break in slope with groundwater inputs to slope wetlands (Brinson
1993).
Some States and Tribes have chosen to refine the spatial resolution of Omernik's ecoregional
boundaries for management of aquatic resources (e.g., Region 3 and Florida). For example, the
37
-------
December 2006- DRAFT Chapter 3. Classification of Wetlands
1417 State of Florida has defined subecoregions for streams based on analysis of macroinvertebrate
1418 data from 100 minimally-impacted sites. Efforts are currently underway to define ecoregions for
1419 Florida wetlands based on variables influencing the water budget and plant community
1420 composition (Dougherty et al. 2000, Lane 2000).
1421
1422 ENVIRONMENTALLY-BASED CLASSIFICATION SYSTEMS
1423
1424 Wetland habitat types are described very simply but coarsely by Shaw and Fredine (1956,
1425 Circular 39), ranging from temporarily-flooded systems to ponds. A more refined hierarchical
1426 classification system is available based on vegetation associations; for example the system
1427 developed by the Nature Conservancy for terrestrial vegetation that includes some wetland types
1428 (Grossman et al. 1998). Vegetation associations have also been used to classify Great Lakes
1429 coastal wetlands within coastal geomorphic type (Michigan Natural Features Inventory 1997).
1430
1431 COWARDIN CLASSIFICATION SYSTEM
1432
1433 The Cowardin classification system (Cowardin et al. 1979) was developed for the U.S. Fish and
1434 Wildlife Service (FWS) as a basis for identifying, classifying, and mapping wetlands, special
1435 aquatic sites, and deepwater aquatic habitats. The Cowardin system combines a number of
1436 approaches incorporating landscape position, hydrologic regime and habitat (vegetative) type
1437 (http://www.nwi.fws.gov) (Figure 3.7). Wetlands are categorized first by landscape position
1438 (tidal, riverine, lacustrine, and palustrine), then by cover type (e.g., open water, submerged
1439 aquatic bed, persistent emergent vegetation, shrub wetlands, and forested wetlands), and then by
1440 hydrologic regime (ranging from saturated or temporarily-flooded to permanently flooded).
1441 Modifiers can be added for different salinity or acidity classes, soil type (organic vs. mineral), or
1442 disturbance activities (impoundment, beaver activity). Thus, the Cowardin system includes a
1443 mixture of geographically-based factors, proximal forcing functions (hydrologic regime, acidity),
1444 anthropogenic disturbance regimes, and vegetative outcomes. In practice, the Cowardin system
1445 can be aggregated by combination of hydrogeomorphic (HGM) type and predominant vegetation
1446 cover if digital coverages are available (Ernst et al. 1995).
1447
1448 HYDROGEOMORPHIC CLASSIFICATION SYSTEM(S)
1449
1450 Brinson (1993) has defined a hydrogeomorphic classification system for wetlands, based on
1451 geomorphic setting, dominant water source (Figure 3.4), and dominant hydrodynamics (Figure
1452 3.5; http://www.wes.army.mil/el/wetlands). Seven classes have been described: depressional,
1453 lacustrine fringe, tidal fringe, slope, riverine, mineral soil flats, and organic soil flats (Smith et al.
1454 1995). Also see Hydrogeomorphic Classification in
1455 http://www.epa.gov/waterscience/criteria/wetlands/7Classification.pdf
1456
1457 Depressional systems, as the name implies, are located in topographic depressions where surface
1458 water can accumulate. Depression wetlands can be further classified based on presence of inlets
38
-------
December 2006- DRAFT Chapter 3. Classification of Wetlands
1459 or outlets and primary water source as closed, open/groundwater, or open/surface water
1460 subclasses.
1461
1462 Lacustrine fringe wetlands are located along lake shores where the water elevation of the lake
1463 determines the water table of the adjacent wetland. Great Lakes coastal wetlands represent one
1464 important region of lacustrine fringe wetlands. These coastal systems are strongly influenced by
1465 coastal forming processes, and, as such, have been further classified by geomorphic type through
1466 various schemes (Jaworski and Raphael 1979, and others summarized in Michigan Natural
1467 Features Inventory 1997). These geomorphic coastal positions will further influence the
1468 predominant source of water and the degree and type of energy regime (riverine vs. seiche and
1469 wave activity). Tidal fringe wetlands occupy a similar position relative to marine coasts and
1470 estuaries, where water level is influenced by sea level. Tidal fringe wetlands can be broken down
1471 further based on salinity into euhaline vs. mixohaline subclasses. Slope wetlands occur on slopes
1472 where groundwater discharges to the land surface but typically do not have the capacity for
1473 surface water storage (Figure 3.6). Riverine wetlands are found in floodplains and riparian zones
1474 associated with stream channels. Riverine systems can be broken down based on watershed
1475 position (and thus hydrologic regime) into tidal, lower perennial, upper perennial, and
1476 nonperennial subclasses. Mineral soil flats are in areas of low topographic relief (e.g.,
1477 interfluves, relic lake bottoms, and large floodplain terraces) with precipitation as the main
1478 source of water. The topography of organic soil flats (e.g., peatlands), in contrast, is controlled
1479 by the vertical accretion of organic matter.
1480
1481 The HGM classification system is being further refined to the subclass level for different regions
1482 or states and classes (Cole et al. 1997, http://www.wes.army.mil/el/wetlands). In addition to the
1483 classification factors described above, Clairain (2002) suggests using parameters such as the
1484 degree of connection between the wetland and other surface waters (depressional wetlands),
1485 salinity gradients (tidal), degree of slope or channel gradient (slope and riverine wetlands),
i486 position in the landscape (riverine, slope), and a scaling factor (stream order, watershed size or
1487 floodplain width for riverine subclasses). In some cases, existing regional schemes have been
1488 used as the basis for subclass definition (e.g., Stewart and Kantrud 1971, Golet and Larson 1974,
1489 Wharton et al. 1982, Weakley and Schafale 1991, Keough et al. 1999).
1490
1491 The HGM classification system has been applied primarily to assess wetland functions related to
1492 hydrology, biological productivity, biogeochemical cycling, and habitat (Smith et al. 1995,
1493 http://www.wes.army.mil/el/wetlands/pdfs/wrpde9.pdf). The same environmental parameters
1494 that influence wetland functions also determine hydrologic characteristics and background water
1495 quality, which in turn drive wetland habitat structure and community composition, and the
1496 timing of biotic events. Thus, the HGM classification system can serve as a basis for partitioning
1497 variability in reference trophic status and biological condition, as well as defining temporal
1498 strategies for sampling.
1499
1500
39
-------
December 2006- DRAFT Chapter 3. Classification of Wetlands
1501 COMPARISON OF ENVIRONMENTALLY-BASED CLASSIFICATION SYSTEMS
1502
1503 If an integrated assessment of aquatic resources within a watershed or region is desired, it may
1504 be useful to consider intercomparability of classification schemes for wetlands, lakes, and
1505 riverine systems to promote cost-effective sampling and ease of interpretation. The HGM
1506 approach could integrate readily with a finer level of classification for lake type because lentic
1507 systems are separated out as lacustrine fringe or depressional wetlands based on lake or pond
1508 size and influence of water level on the adjacent wetland. Lacustrine classification systems for
1509 water quality have included geography (climate + bedrock characteristics, Gorham et al. 1983)
1510 or hydrologic setting (Winter 1977, Eilers et al. 1983) as factors for categorization. McKee et al.
1511 (1992) suggest a modification of Cowardin's system, for Great Lakes coastal wetlands
1512 incorporating landscape position (system), depth zone (littoral vs. limnetic subsystems),
1513 vegetative or substrate cover (class and subclass), and modifiers of ecoregions, water level
1514 regimes, fish community structure, geomorphic structure, and human modification. In contrast,
1515 the Michigan Natural Features Inventory (1997) categorizes Great Lakes coastal wetlands by
1516 Great Lake, then nine unique geomorphic types within lakes, then vegetative association.
1517
1518 For lotic systems, Brinson et al. (1995) describes an approach to further classify riverine classes
1519 into subclasses based on watershed position and stream size/permanence. This strategy is
1520 consistent with current monitoring efforts to develop stream IBIs (Indices of Biotic Integrity),
1521 which typically use stream order as a surrogate for watershed size in explaining additional
1522 background variation in IBI scores (USEPA 1996). A more detailed classification of stream
1523 reach types, based on hydrogeomorphic character, is described by Rosgen (1996). This
1524 classification scheme has been predominantly applied to assessments of channel stability and
1525 restoration options, and not to development of criteria. Gephardt et al. (1990) described a cross-
1526 walk between riparian and wetland classification and description procedures.
1527
40
-------
December 2006- DRAFT
Chapter 3. Classification of Wetlands
DPI 1NO
I t
s S
M
S »
'. 3
£ IE
S f
EiUUfilt EW04IO
d ! JB1IL.A1
•EHWS —
1528
1529
1530
1531
UH*»0 PAluSTdNf
I THI-l'iM*!-:!. '•'!;.• Jl'l',
b SF(lSl«BHY flOUWll
[ SE1W>F*«AWN-[.T FLOODED
i iNie»MTT(«Tl»
i crhMA»r.ftiii.v n.
1 ... Hi', ii •
LOW HAltl
Figure 3.7. (Top) Cowardin hierarchy of habitat types for estuarine systems;
(Bottom) Palustrine systems, from Cowardin et al. 1979.
41
-------
December 2006- DRAFT Chapter 3. Classification of Wetlands
1532 COMBINATIONS OF GEOGRAPHIC AND ENVIRONMENTALLY-BASED APPROACHES
1533
1534 It is possible to combine geographically-based classification with hydrogeomorphic and/or
1535 habitat-based approaches. For example, a scheme could be defined that nests Cowardin
1536 (Cowardin et al. 1979) vegetative cover class within HGM class within ecoregion. Maxwell et al.
1537 (1995) have defined a scheme for linking geographically-based units based on geoclimatic
1538 setting (domains => divisions => provinces => sections => subsections) to watersheds and
1539 subwatersheds , and thus to riverine systems composed of valley segments, stream reaches, and
1540 channel units, or to lacustrine systems composed of lakes, lake depth zones, and lake
1541 sites/habitat types.
1542
1543 Maxwell et al. (1995) also define a series of fundamental hydrogeomorphic criteria for
1544 classifying wetlands based on Brinson (1993) and Winter (1992), including physiography
1545 (landscape position), water source, hydrodynamics, and climate. The first three of these are
1546 similar to the HGM classification system, whereas moisture regimes and soil temperature
1547 regimes are generally consistent at the province level (see summary tables in Keys et al. 1995).
1548 Finer scale variation in landforms is captured at the level of sections and below, which in turn
1549 will determine the dominance of different hydrogeomorphic classes of wetlands and associated
1550 surface waters (lakes and rivers). Characteristics and relative advantages and disadvantages of
1551 different classification systems are summarized in Table 2.
1552
1553
1554 3.3 SOURCES OF INFORMATION FOR MAPPING WETLAND CLASSES
1555
1556 In order to select wetlands for sampling in a random- or random-stratified design (described in
1557 Chapter 4), it is important to have a record of wetland locations to choose from, preferably
1558 categorized by the classification system of interest. For some, but not all portions of the country,
1559 wetlands have been mapped from aerial photography through the National Wetlands Inventory
1560 (NWI) maintained by the U.S. Fish and Wildlife Service (http://www.fws.gov/nwi/: Dahl 2005).
1561 In other cases, individual states have developed inventories, or researchers have developed lists
1562 for specific types of wetlands within a given region, e.g., Great Lakes coastal wetlands
1563 (Herdendorf et al. 1981). In order to sample these mapped wetland areas in a random fashion, it
1564 is important to have a list of each wetland that occurs within each class and its associated area. A
1565 geographic information system (GIS) allows one to automatically produce a list of all wetland
1566 polygons by type within a specified geographic region. Sources of digital information for
1567 mapping and/or classifying wetlands in a GIS are presented in the Land-Use Characterization for
1568 Nutrient and Sediment Risk Assessment Module
1569 (http://www.epa.gov/waterscience/criteria/wetlands/17LandUse.pdf). In areas for which digital
1570 NWI maps do not yet exist, potential wetland areas can be mapped using GIS tools
42
-------
December 2006- DRAFT
Chapter 3. Classification of Wetlands
1571
Table 2. Comparison of landscape and wetland classification schemes.
Classification
scheme
Bailey's ecoregions
Omernik ecoregions
Ecological units
(Maxwell et al. 1995)
USAGE
Hydrogeomorphic
Classes
Rosgen channel
types
Scale
Nationwide
Nationwide
Nationwide
Nationwide at
class level;
regionalized at
subclass level
Nationwide
Hierarchical?
Yes
No
Yes
Yes - limited
Yes
Levels of strata
Domains
Divisions
Provinces
Sections
Ecoregions
Subecoregions
Domain
Divisions
Provinces
Sections
Subsections
Class
Subclass
Level I
Level II
Advantages
Only natural
attributes
included
Digital maps
Digital maps
Digital maps
Specific for
wetlands
Captures
differences
in hydrologic
regime for
riverine
wetlands
Disadvantages
Terrestrial basis
Untested for wetlands
No hydrology
Combines land-use
with natural
attributes
Untested for most
wetlands
No hydrology
Greater number of
strata and units
than for ecoregions
Untested for wetlands
Subclasses not
comparable across
different regions
More focused on
instream channel
form than riparian
characteristics
Riverine only
Not mapped
Potential links with other schemes
Could form first strata for any of the
schemes below ecological units
Could form first strata for any of the
schemes below ecological units
Could form first strata for any of the
schemes below ecological units
Ties to classification schemes
already defined within
hydrogeomorphic types
Intermediate strata between
geographic and habitat-scale
Intermediate strata between hydro-
geomorphic type and habitat-
scale
43
-------
December 2006- DRAFT
Chapter 3. Classification of Wetlands
Table 2. Comparison of landscape and wetland classification schemes.
Classification
scheme
Anderson land-cover
classes
Circular 39 classes
National Wetland
Inventory
Vegetation
associations
Scale
Nationwide
Nationwide
Nationwide
International
Hierarchical?
Yes
No
Yes
Yes
Levels of strata
Level 1
Level II
Level III
Class
System
Subsystem
Class
Subclass
Hydrologic
modifier
Other modifiers
System
Formation class
Formation
subclass
Formation group
Formation
subgroup
Formation
alliance
Association
Advantages
Common basis
for land-
use/land-
cover
mapping
Popular
recognition
Digital maps
available for
much of
nation (but
smallest
wetlands
omitted)
Consistency
across
terrestrial
and aquatic
systems
Disadvantages
Not functionally based
Mixture of criteria used
to distinguish
classes
Not mapped
Inconsistencies in
mapping water
quality modifiers
Limited consideration
of
hydrogeomorphic
type
Not functionally based
No digital maps
Taxa specific
Potential links with other schemes
Cross-walk w NWI system possible
Strata below geographic but
contains mixture of
hydrogeomorphic type and
habitat type
Strata below geographic
Hydrogeomorphic class could be
improved by linkw HGM system
Could be used as lowest level within
other schemes
1572
44
-------
December 2006- DRAFT Chapter 3. Classification of Wetlands
1573 to predict relative wetness (e.g., Phillips 1990) or soil survey maps with hydric soil series can be
1574 used. It should be noted that in areas in which hydrology has been significantly altered
1575 (e.g.,through ditching, tiling, or construction of urban stormwater systems), areas of potential
1576 wetlands could have been removed already. Similarly, although there are no current maps of
1577 wetlands by hydrogeomorphic class, these could be derived through GIS techniques using a
1578 combination of wetland coverages, hydrography (adjacency to large lakes and rivers), and digital
1579 elevation models to derive landforms (mineral and organic soil flats) and/or landscape position
1580 (slope and depressional wetlands).
1581
1582
1583 3.4 DIFFERENCES IN NUTRIENT REFERENCE CONDITION OR SENSITIVITY
1584 TO NUTRIENTS AMONG WETLAND CLASSES
1585
1586 Very few studies to verify classification systems for wetland nutrient monitoring have been
1587 completed, although a number of monitoring strategies have been implemented based on pre-
1588 selected strata. Monitoring efforts to develop or assess biological criteria generally have used a
1589 combination of geographic region and hydrogeomorphic class or subclass (e.g.,Cole et al. 1997,
1590 Bennett 1999, Apfelbeck 1999, Michigan Natural Features Inventory 1997). Analysis of plant
1591 associations has been used to derive empirical classifications based on factors such as landscape
1592 position, water source, climate, bedrock, and sediment hydraulic conductivity (Weakley and
1593 Schafale 1991, Nicholson 1995, Halsey et al. 1997, Michigan Natural Features Inventory 1997).
1594 Only one case of classification based on wetland macroinvertebrate composition was found. For
1595 Australian wetlands, wetland classes grouped by macroinvertebrate communities were
1596 distinguished by water chemistry extremes (low pH, high salinity), degree of nutrient
1597 enrichment, and water color (Growns et al. 1992).
1598
1599 In some cases (e.g., northern peatlands) classification criteria derived on the basis of plant
1600 associations are less powerful in discriminating among nutrient regimes (e.g.,Nicholson 1995);
1601 this may be particularly true where variation in vegetation type is related to differences in major
1602 ion chemistry and pH rather than nutrients. The same is true in southern pocosins, where short
1603 and tall pocosins differ in seasonal hydrology but not soil chemistry. However, when contrasting
1604 pocosins and swamp forests, soil nutrients differed strongly (Bridgham and Richardson 1993).
1605 For some potential indicators of nutrient status such as vegetation N:P ratios, indicator
1606 thresholds will be consistent across species (Koerselman and Meuleman 1996), while response
1607 thresholds for other indicators of plant nutrient status vary across functional plant groupings with
1608 different life history strategies. These differences may indicate potential differences in sensitivity
1609 to excess nutrient loading (McJannet et al. 1995). Thus, vegetation community types are not
1610 always a good predictor of background nutrient concentrations (reference condition) or
1611 sensitivity to nutrient loading.
1612
1613 Sensitivity to nutrient loading (as evidenced by differences in nutrient cycling and availability)
1614 may also be related to differences in hydroperiod among wetlands. Wetland mesocosms exposed
45
-------
December 2006- DRAFT Chapter 3. Classification of Wetlands
1615 to pulse discharges had higher nutrient loss from the water column than those exposed to
1616 continuous flow regimes (Busnardo et al. 1992). Depending on the predominant mechanism for
1617 nutrient loss (e.g., plant uptake versus denitrification), nutrient-controlled primary production
1618 could be either stimulated or reduced. Mineralization rates of carbon, nitrogen, and phosphorus
1619 differ significantly among soils from northern Minnesota wetlands, related to an ombrotrophic to
1620 minerotrophic gradient (i.e., degree of groundwater influence) and aeration status (Bridgham et
1621 al. 1998).
1622
1623 In general, very few definitive tests of alternative classification schemes for wetlands are
1624 available with respect to describing reference condition for either nutrient criteria or biocriteria.
1625 However, evidence from the literature suggests that in many cases, both geographic factors (e.g.,
1626 climate, geologic setting) and landscape setting (hydrogeomorphic type) are expected to affect
1627 both water quality and biotic communities.
1628
1629
1630 3.5 RECOMMENDATIONS
1631
1632 Classification strategies for nutrient criteria development should incorporate factors affecting
1633 background nutrient levels and wetland sensitivity to nutrient loading at several spatial scales.
1634
1635 • Classification of physiographic regions eliminates background variation in lithology and
1636 soil texture (affecting background nutrient levels and sorption capacity), in climate
1637 (affecting seasonality, productivity, decomposition and peat formation), and in
1638 landforms, which determines the predominance of different hydrogeomorphic classes.
1639
1640 • Classification by hydrogeomorphic class reduces background variation in predominant
1641 water and nutrient sources, water depth and dynamics, hydraulic retention time,
1642 assimilative capacity, and interactions with other surface water types (Table 3).
1643
1644 • Classification by water depth and duration (which may or may not be incorporated into
1645 hydrogeomorphic classes) helps to explain variation in internal nutrient cycling,
1646 dissolved oxygen level and variation, and the ability of wetlands to support some higher
1647 trophic levels such as fish and amphibians.
1648
1649 • Classification by vegetation type or zone, whether to inform site selection or to determine
1650 sampling strata within a site helps to explain background variation in predominant
1651 primary producer form (which will affect endpoint selection), as well as turnover rate and
1652 growth rates (which will affect rapidity of response to nutrient loadings).
1653
1654 In general, the choice of specific alternatives among the classification schemes listed above
1655 depends both on their intrinsic value as well as practical considerations, e.g., whether a
1656 classification scheme is available in mapped digital form or can be readily derived from
46
-------
December 2006- DRAFT Chapter 3. Classification of Wetlands
1657 existing map layers, whether a hydrogeomorphic or other classification scheme has been
1658 refined for a particular region and wetland type, and whether classification schemes are
1659 already in use for monitoring and assessment of other waterbody types in a state or region.
47
-------
December 2006- DRAFT
Chapter 3. Classification of Wetlands
Table 3. Features of the major hydrogeomorphic classes of wetlands that may influence background nutrient concentrations,
sensitivity to nutrient loading, nutrient storage forms and assimilative capacity, designated use and choice of endpoints.
HGM Class
Predominant
Nutrient Source(S)
Landscape
Position
Hydrologic Regime
Hydraulic
Retention Time
Nutrient
Assimilation
Capacity
Organic Flats
Atmospheric
Deposition
Saturated, Little
Standing Water
Decades
Low
Mineral Flats
Atmospheric
Deposition,
Groundwater
Saturated, Little
Standing Water
Decades
High Sorption
Capacity
Depressional
Runoff (Particulate
And Dissolved),
Surface And
Groundwater
Depth And
Duration Vary
From Saturated To
Temporary To
Seasonal To
Semi-Permanent
To Permanent
Inundation
Varies With
Inflows/Outflows,
Landscape
Position
High Sorption,
Plant Uptake,
(Limited) Sediment
Storage
Riverine
Runoff
(Particulate),
Overbank Flooding
(Particulate,
Dissolved)
Adjacent To Rivers
Depth, Duration
Vary W River
Flooding Regime
-------
December 2006- DRAFT
Chapter 3. Classification of Wetlands
Table 3 cont'd
HGM Class
Predominant
Vegetation Growth
Form
Top Trophic Level
Commercially-
Important
Fish/Wildlife
Recreational Use
Likely
Drinking Water
Source
Downstream
Organic Flats
Mosses
Sedges
Mammals
Birds
Amphibians
Invertebrates
Mineral Flats
Sedges
Mammals
Birds
Amphibians
Invertebrates
Depressional
Varies With Zone
And Duration Of
Flooding:
Wooded
Grass/Sedge
Emergents
Submerged
Aquatics*
Mammals
Birds
Mudminnows
Amphibians
Invertebrates
Waterfowl
Yes
Possible
Riverine
Wooded,
Emergent
Vegetation
Submerged
Aquatics*
Fish
Birds
Mammals
Fish*
Yes
Likely
Fringe
Varies With Zone:
Grass/Sedge
Emergents
Submerged
Aquatics*
Fish
Birds
Mammals
Waterfowl
Fish*
Yes
Possible
Slope
Wooded
Grasses
Sedges
Mammals
Birds
Amphibians
Invertebrates
1660
49
-------
1661
December 2006- DRAFT Chapter 4. Sampling Design of Wetland Monitoring
Chapter 4 Sampling Design for Wetland Monitoring
1662
1663 4.1 INTRODUCTION
1664
1665 This chapter provides technical guidance on designing effective sampling programs for State and
1666 Tribal wetland water quality monitoring programs. EPA recommends that States and Tribes
1667 begin wetland monitoring programs to collect water quality and biological data in order to
1668 characterize the condition of existing wetlands as they develop nutrient criteria that protect their
1669 wetlands. The best monitoring programs are designed to assess wetland conditions with
1670 statistical rigor while maximizing available resources.
1671
1672 At the broadest level, monitoring data should:
1673
1674 1. Detect and characterize the condition of existing wetlands.
1675
1676 2. Describe whether wetland conditions are improving, degrading, or staying the same.
1677
1678 3. Define seasonal patterns, impairments, deviations in status in wetland conditions.
1679
1680 Water quality monitoring programs should collect a sufficient number of samples over time and
1681 space to identify changes in system condition or estimate average conditions with statistical
1682 rigor. Three approaches to study design for assessing water quality, biological and ecological
1683 condition, as well as identifying degradation in wetlands, are described in this chapter. Specific
1684 issues to consider in designing monitoring programs for wetland systems are also discussed in
1685 this chapter. The study designs presented here can be tailored to fit the goals of specific
1686 monitoring programs.
1687
1688 The three approaches described below (Section 4.3) (probabilistic sampling, targeted/tiered, and
1689 Before/After - Control/Impact [BACI]), present study designs that allow one to obtain a
1690 significant amount of information with relatively minimal effort. Probabilistic sampling begins
1691 with a large-scale random monitoring design that is reduced as the wetland system conditions are
1692 characterized. This approach is used to find the average condition of each wetland class in a
1693 specific region. Probabilistic sampling design is frequently used for new large scale monitoring
1694 programs at the State and Federal level (e.g., Environmental Monitoring and Assessment
1695 Program (EMAP), Regional Environmental Monitoring and Assessment Program (REMAP),
1696 State programs [e.g., Maine, Montana, Wisconsin]). The tiered or targeted approach to
1697 monitoring begins with coarse screening and proceeds to more detailed monitoring protocols as
1698 impaired and high-risk systems are identified and targeted for further investigation. Targeted
1699 sampling design provides a triage approach to more thoroughly assess condition and diagnose
1700 stressors in wetland systems in need of restoration, protection and intensive management.
1701 Several State pilot projects use this method or a modification of this method for wetland
50
-------
December 2006- DRAFT Chapter 4. Sampling Design of Wetland Monitoring
1702 assessment (e.g., Florida, Ohio, Oregon, Minnesota). The synoptic approach described in
1703 Kentula et al. (1993) uses a modified targeted sampling design. The BACI design and its
1704 modifications are frequently used to assess the success of restoration efforts or other
1705 management experiments such as those describe in Case Study 2 in Appendix B2. BACI design
1706 allows for comparisons in similar systems over time to determine the rate of change in relation to
1707 the management activity, e.g., to assess the success of a wetland hydrologic restoration.
1708 Detenbeck et al. (1996) used BACI design for monitoring water quality of wetlands in the
1709 Minneapolis/St. Paul, Minnesota metro area.
1710
1711 Monitoring programs should be designed to describe what the current conditions are and to
1712 answer under what conditions impairment may occur. A well-designed monitoring program can
1713 contribute to determining those conditions.
1714
1715 Sampling design is dependent on the management question being asked. Sampling efforts
1716 should be designed to collect information that will answer the management question. For
1717 example, probabilistic sampling might be good for ambient (synoptic) monitoring programs,
1718 BACI for evaluating management actions such as restoration, and targeted sampling/stratified
1719 and random sampling for developing index of biotic integrity (IBIs) or nutrient criteria
1720 thresholds. In practice, some state programs likely will need to use a combination of approaches.
1721
1722
51
-------
December 2006- DRAFT Chapter 4. Sampling Design of Wetland Monitoring
1723 4.2 CONSIDERATIONS FOR SAMPLING DESIGN
1724
1725 DESCRIBING THE MANAGEMENT QUESTION
1726
1727 Clearly defining the question being asked (identifying the hypothesis) encourages the use of
1728 appropriate statistical analyses, reduces the occurrence of Type I (false positive) errors, and
1729 increases the efficient use of management resources (Suter 1993; Leibowitz et al., 1992; Kentula
1730 et al., 1993). Beginning a study or monitoring program with carefully defined questions and
1731 objectives helps to identify the statistical analyses most appropriate for the study, and reduces
1732 the chance that statistical assumptions will be violated. Management resources are optimized
1733 because resources are directed at monitoring that which is most likely to answer management
1734 questions. In addition, defining the specific hypotheses to be tested, carefully selecting reference
1735 sites, and identifying the most useful sampling interval can help reduce the uncertainty
1736 associated with the results of any sampling design, and further conserve management resources
1737 (Kentula et al., 1993). Protecting or improving the quality of a wetland system often depends on
1738 the ability of the monitoring program to identify cause-response relationships, for example, the
1739 relationship of nutrient concentration (causal variable) to nutrient content of vegetation or
1740 vegetation biomass (response variable). Cause-response relationships can be identified using
1741 large sample sizes, and systems that span the gradient (low to high) of wetland quality. All
1742 ranges of response should be observed along the causal gradient from minimally disturbed to
1743 high levels of human disturbance.
1744
1745 Monitoring efforts often are prioritized to best utilize limited resources. For example, the Oregon
1746 case study chose not to monitor depressional wetlands due to funding constraints. They further
1747 tested the degree of independence of selected sites (and thus the need to monitor all of those
1748 sites) using cluster analysis and other statistical tests
1749 (http://www.epa.gov/owow/wetlands/bawwg/case/or.html). Frequency of monitoring should be
1750 determined by the management question being asked, and the intensity of monitoring necessary
1751 to collect enough information to answer the question. In addition, monitoring should identify the
1752 watershed level activities that are likely to result in ecological degradation of wetland systems
1753 (Suter etal. 1993).
1754
1755 SITE SELECTION
1756
1757 Site selection is one of many important tasks in developing a monitoring program (Kentula et al.
1758 1993). Site selection for a monitoring program is based on the need to sample a sufficiently large
1759 number of wetlands to establish the range of wetland quality in a specific regional setting.
1760 Wetland monitoring frequently includes an analysis of both watershed/landscape characteristics
1761 and wetland specific characteristics (Kentula et al.,1993; Leibowitz et al., 1992). Therefore,
1762 wetland sampling sites should be selected based on land use in the region so that watersheds
1763 range from minimally impaired with few expected stressors to high levels of development (e.g.,
1764 agriculture, forestry, or urban) with multiple expected stressors (see the Land-Use
52
-------
December 2006- DRAFT Chapter 4. Sampling Design of Wetland Monitoring
1765 Characterization for Nutrient and Sediment Risk Assessment, Wetland module #17). There is
1766 often a lag in time between the causal stress and the response in the wetland system. This time
1767 lag between stress and response and the duration of this lag depends on many factors including
1768 the type of stressor, climate, and system hydrology; these factors should be considered when
1769 selecting sites to establish the range of wetland quality within a region.
1770
1771 LANDSCAPE CHARACTERIZATION
1772
1773 The synoptic approach described in Liebowitz et al. (1992) provides a method of rapid
1774 assessment of wetlands at the regional and watershed level that can help identify the range of
1775 wetland quality within a region. Liebowitz et al. (1992) recommend an initial assessment for site
1776 selection based on current knowledge of watershed and landscape level features; modification of
1777 such an assessment can be made as more data are collected. Assessing watershed characteristics
1778 through aerial photography and the use of geographical information systems (GIS) linked to
1779 natural resource and land-use databases, can aid in identifying reference and degraded systems
1780 (see the Land-Use Characterization for Nutrient and Sediment Risk Assessment, Wetland
1781 module #17); Johnston et al., 1988, 1990; Gwin et al., 1999; Palik et al., 2000; Brown and Vivas
1782 2004). Some examples of watershed characteristics which can be evaluated using GIS and aerial
1783 photography include land use, land cover (including riparian vegetation), soils, bedrock,
1784 hydrography, and infrastructure (e.g., roads or railroads). Changes in point sources can be
1785 monitored through the NPDES permit program (USEPA 2000). Changes in nonpoint sources can
1786 be evaluated through the identification and tracking of wetland loss and/or degradation,
1787 increased residential development, urbanization, increased tree harvesting, shifts to more
1788 intensive agriculture with greater fertilizer use or increases in livestock numbers, and other land
1789 use changes. Local planning agencies should be informed of the risk of increased anthropogenic
1790 stress and encouraged to guide development accordingly.
1791
1792 IDENTIFYING AND CHARACTERIZING REFERENCE WETLANDS
1793
1794 The term "reference" in this document refers to those systems that are least impaired by
1795 anthropogenic effects. The use of the term reference is confusing because of the different
1796 meanings that are currently in use in different classification methods, particularly its use in
1797 hydrogeomorphic [HGM] wetland classification. A discussion of the term reference and its
1798 multiple meanings is provided in Chapter 3.
1799
1800 Watersheds with little or no development that receive minimal anthropogenic inputs could
1801 potentially contain wetlands that may serve as minimally impaired reference sites. Watersheds
1802 with a high percentage of the drainage basin occupied by urban areas, agricultural land, and
1803 altered hydrology are likely to contain wetlands that are impaired or could potentially be
1804 considered "at risk" for developing problems. Wetland loss in the landscape also should be
1805 considered when assessing watershed characteristics for reference wetland identification.
1806 Biodiversity can become impoverished due to wetland fragmentation or decreases in regional
53
-------
December 2006- DRAFT Chapter 4. Sampling Design of Wetland Monitoring
1807 wetland density even in the absence of site-specific land-use activities. Reference wetlands may
1808 be more difficult to locate if fragmentation of wetland habitats is significant, and may no longer
1809 represent the biodiversity of minimally disturbed wetlands in the region. The continued high rate
1810 of wetland loss in most States and Tribal lands dictates that multiple reference sites be selected
1811 to insure some consistency in reference sites for multiple year sampling programs (Liebowitz et
1812 al., 1992; Kentula et al., 1993). Once the watershed level has been considered, a more site-
1813 specific investigation can be initiated to better assess wetland condition.
1814
1815 The ideal reference site will have similar soils, vegetation, hydrologic regime and landscape
1816 setting to other wetlands in the region (Adamus 1992; Liebowitz et al., 1992; Kentula et al.,
1817 1993; Detenbeck et al., 1996). Classification of wetlands, as discussed in Chapter 3, may aid in
1818 identifying appropriate reference wetlands for specific regions and wetland types. Wetland
1819 classification should be supplemented with information on wetland hydroperiod to assure that
1820 the selected reference wetlands are truly representative of wetlands in the region, class or
1821 subclass of interest. Reference wetlands may not be available for all wetland classes. In this case,
1822 data from systems that are as close as possible to the assumed unimpaired state of wetlands in the
1823 wetland class of interest should be sought from States or Tribes within the same geologic
1824 province. Development of a conceptual reference may be important, if appropriate reference sites
1825 cannot be found in the local region or geologic province. Techniques for defining a conceptual
1826 reference are discussed at some length in Harris et al. (1995), Trexler (1995), and Toth et al.
1827 (1995).
1828
1829 Reference wetlands should be selected based on low levels of human alteration in their
1830 watersheds (Liebowitz et al. 1992; Kentula et al. 1993; USEPA 2000). Selecting reference
1831 wetlands usually involves assessment of land-use within watersheds, and visits to individual
1832 wetland systems to ground-truth expected land-use and check for unsuspected impacts. Ground-
1833 truthing visits to reference wetlands are crucial for identification of ecological impairment that
1834 may not be apparent from land-use and local habitat conditions. Again, sufficient sample size is
1835 important to characterize the range of conditions that can be expected in the least impacted
1836 systems of the region (Detenbeck et al. 1996). Reference wetlands should be identified for each
1837 ecoregion or geological province in the State or Tribal lands and then characterized with respect
1838 to ecological integrity. A minimum of three low impact reference systems is recommended for
1839 each wetland class for statistical analyses. However, power analysis can be performed to
1840 determine the degree of replication necessary to detect an impact to the systems being
1841 investigated (Detenbeck et al. 1996; Urquhart et al. 1998). Highest priority should be given to
1842 identifying reference systems for those wetland types considered to be at the greatest risk from
1843 anthropogenic stress.
1844
1845 WHEN TO SAMPLE
1846
1847 Sampling may be targeted to the periods when effects are most likely to be detected - the index
1848 period. The appropriate index period should be defined by what the investigator is trying to
54
-------
December 2006- DRAFT Chapter 4. Sampling Design of Wetland Monitoring
1849 investigate, and what taxonomic assemblage or parameters are being used for that investigation
1850 (Barbour et al. 1999). For example, increased nutrient concentrations and sedimentation from
1851 non-point sources may occur following periods of high runoff during spring and fall, while point
1852 sources of nutrient pollutants may cause plankton blooms and/or increased water and soil
1853 nutrient concentrations in wetland pools during times of low rainfall. Hence, different index
1854 periods may be needed to detect effects from point source and non-point source nutrients,
1855 respectively. Each taxonomic assemblage studied also should have an appropriate index period -
1856 usually in the growing season (see assemblage methods in the Maine case study:
1857 http ://www. epa. gov/waterscience/criteria/wetlands/index.html).
1858
1859 The index period window may be early in the growing season for amphibians and algae. Other
1860 assemblages, such as vegetation and birds, may benefit from a different sampling window for the
1861 index period; see the assemblage specific modules for recommendations. Once wetland
1862 condition has been characterized, one-time annual sampling during the appropriate index period
1863 may be adequate for multiple year monitoring of indicators of nutrient status, designated use, and
1864 biotic integrity. However, criteria and ecological indicator development may benefit from more
1865 frequent sampling to define conditions that relate to the stressor or perturbor of interest (Karr and
1866 Chu 1999; Stevenson 1996; Stevenson 1997). Regardless of the frequency of sampling, selection
1867 of index periods and critical review of the data gathered and analyzed should be done to
1868 scientifically validate the site characterization and index periods for data collection.
1869
1870 Ideally, water quality monitoring programs produce long-term data sets compiled over multiple
1871 years, to capture the natural, seasonal and year-to-year variations in biological communities and
1872 waterbody constituent concentrations (e.g., Tate 1990; Dodds et al. 1997; McCormick et al.
1873 1999; Craft 2001; Craft et al., 2003; Zheng et al., 2004). Multiple-year data sets can be analyzed
1874 with statistical rigor to identify the effects of seasonality and variable hydrology. Once the
1875 pattern of natural variation has been described, the data can be analyzed to determine the
1876 ecological state of the waterbody. Long-term data sets have also been important in influencing
1877 management decisions about wetlands, most notably in the Everglades, where long-term data
1878 sets have induced Federal, State, and Tribal actions for conservation and restoration of the
1879 largest wetland system in the US (see Davis and Ogden 1994; Everglades Interim Report, South
1880 Florida Water Management District [SFWMD, 1999]; Everglades Consolidated Report
1881 [SFWMD, 2000, 2001]; 1994 Everglades Forever Act, Florida Statute § 373.4592).
1882
1883 In spite of the documented value of long-term data sets, there is a tendency to intensively study a
1884 waterbody for one year before and one year after treatment. A more cost-effective approach may
1885 be to measure only the indices most directly related to the stressor of interest (i.e., those
1886 parameters or indicators that provide the best information to answer the specific management
1887 question), but to double or triple the monitoring period. Multiple years (two or more) of data are
1888 often needed to identify the effects of years with extreme climatic or hydrologic conditions.
1889 Comparisons over time between reference and at risk or degraded systems can help describe
1890 biological response and annual patterns in the presence of changing climatic conditions. Multi-
55
-------
December 2006- DRAFT Chapter 4. Sampling Design of Wetland Monitoring
1891 year data sets also can help describe regional trends. Flooding or drought may significantly
1892 affect wetland biological communities and the concentrations of water column and soil
1893 constituents. Effects of uncommon climatic events can be characterized to discern the overall
1894 effect of management actions (e.g., nutrient reduction, water diversion), if several years of data
1895 are available to identify the long-term trends.
1896
1897 At the very minimum, two years of data before and after specific management actions, but
1898 preferably three or more each, are recommended to evaluate the cost-effectiveness of
1899 management actions with some degree of certainty (USEPA 2000). If funds are limited,
1900 restricting sampling frequency and/or numbers of indices analyzed should be considered to
1901 preserve a longer-term data set. Reducing sampling frequency or numbers of parameters
1902 measured will allow for effectiveness of management approaches to be assessed against the high
1903 annual variability that is common in most wetland systems. Wetlands with high hydrological
1904 variation from year to year may benefit from more years of sampling both before and after
1905 specific management activities to identify the effects of the natural hydrologic variability
1906 (Kadlec and Knight 1996).
1907
1908 CHARACTERIZING PRECISION OF ESTIMATES
1909
1910 Estimates of cause-response relationships, nutrient and biological conditions in reference
1911 systems and wetland conditions in a region are based on sampling, hence precision should be
1912 assessed. Precision is defined as the "measure of the degree of agreement among the replicate
1913 analyses of a sample, usually expressed as the standard deviation" (APHA 1999). Determining
1914 precision of measurements for one-time assessments from single samples in a wetland is often
1915 important. The variation associated with one-time assessments from single samples can be
1916 determined by re-sampling a specific number of wetlands during the survey. Measurement
1917 variation among replicate samples then can be used to establish the expected variation for one-
1918 time assessment of single samples. Re-sampling does not establish the precision of the
1919 assessment process, but rather identifies the precision of an individual measurement (Kentula et
1920 al. 1993).
1921
1922 Re-sampling frequency is often conducted for one wetland site in every block often sites.
1923 However, investigators should adhere to the objectives of re-sampling (often considered an
1924 essential element of QA/QC) to establish an assessment of the variation in a one-time/sample
1925 assessment. Often, more than one in ten samples should be replicated in monitoring programs to
1926 provide a reliable estimate of measurement precision (Barbour et al. 1999). The reader should
1927 understand that this is a very brief description of the concerns about precision, and that any
1928 monitoring program or study involving monitoring should include consultation with a
1929 professional statistician before the program begins and regularly during course of the monitoring
1930 program to assure statistical rigor.
1931
1932
56
-------
December 2006- DRAFT Chapter 4. Sampling Design of Wetland Monitoring
1933 4.3 SAMPLING PROTOCOL
1934
1935 APPROACHES TO SAMPLING DESIGN
1936
1937 The following sections discuss three different approaches to sampling design, probabilistic,
1938 targeted, and BACI. These approaches have advantages and disadvantages that under different
1939 circumstances warrant the choice of one approach over the other (Table 4). The decision as to
1940 the best approach for sample design in a new monitoring program should be made by the water
1941 quality resource manager or management team after carefully considering different approaches.
1942 For example, justification of a dose-response relationship is confounded by lack of
1943 randomization and replication, and should be considered in choosing a sampling design for a
1944 monitoring program.
1945
1946 PROBABILISTIC SAMPLING DESIGN FOR ASSESSING CONDITION
1947
1948 Probabilistic sampling - a sampling process wherein randomness is requisite (Hayek 1994) - can
1949 be used to characterize the status of water quality conditions and biotic integrity in a region's
1950 wetland system. This type of sampling design is used to describe the average conditions of a
1951 wetland population, identify the variability among sampled wetlands, and to help determine the
1952 range of wetland system conditions in a region. Data collected from a probabilistic random
1953 sample design generally will be characteristic of the dominant class or type of wetland in the
1954 region, but rare wetlands may be under-represented or absent from the probabilistically sampled
1955 wetlands. Additional sampling sites may need to be added to precisely characterize the complete
1956 range of wetland conditions and types in the region.
1957
1958 Probabilistic designs are often modified by stratification (such as classification). Stratified
1959 random sampling is a type of probabilistic sampling where a target population is divided into
1960 relatively homogenous groups or classes (strata) prior to sampling based on factors that influence
1961 variability in that population (Hayek 1994). Stratification by wetland size and class or types
1962 ensures more complete information about different types of wetlands within a region. Sample
1963 statistics from random selection alone would be most characteristic of the dominant wetland type
1964 in a region if the population of wetlands is not stratified.
1965
1966 Many state 305(b) and watershed monitoring programs utilize stratified random sampling
1967 designs and we will further discuss this type of probabilistic sampling. Maine, Montana and
1968 Wisconsin pilot projects all use stratified random sampling design. Details of these monitoring
1969 designs can be found in the Case Studies module #14 [APPENDIX B] and can be found on the
1970 web at http://www.epa.gov/waterscience/criteria/wetlands/index.html.
1971
1972 Stratification is based on identifying wetland systems in a region (or watershed) and then
1973 selecting an appropriate sample of systems from the defined population. The determination of an
1974 appropriate sample population usually is dependent on the management questions being asked. A
57
-------
December 2006- DRAFT Chapter 4. Sampling Design of Wetland Monitoring
1975 sample population of isolated depressional wetlands could be identified as a single stratum, but
1976 investigations of these wetlands would not provide any information on riparian wetlands in the
1977 same region. If the goal of the monitoring program is to identify wetland condition for all
1978 wetland classes within a region, then a sample population of wetlands should be randomly
1979 selected from all wetlands within each class. In practice, most State and Tribal programs stratify
1980 random populations by size, wetland class (see chapter 3), and landscape characteristics or
1981 location (see http://www.epa.gov/waterscience/criteria/wetlands/index.html, module #14).
1982
1983 Once the wetlands for each stratum have been identified, the sample population can be stratified
1984 spatially to ensure even spatial coverage of the assessment and indirectly increase the types of
1985 wetlands sampled (assuming classes of wetlands vary spatially). For example, EMAP limits
1986 redundant collection efforts by applying the Generalized Random Tessellation Stratified (GRTS)
1987 design to a map of the area. Sampling sites are chosen by randomly selecting grid cells, and
1988 randomly sampling wetland resources within the chosen grid cells (Paulsen et al. 1991).
1989 Estimates of ecological conditions from these kinds of modified probabilistic sampling designs
1990 can be used to characterize the water quality conditions and biological integrity of wetland
1991 systems in a region, and over time, to distinguish trends in ecological condition within a region.
1992 (See http://www.epa.gov/owow/wetlands/bawwg/case/mtdev.html, and
1993 http://www.epa.gov/owow/wetlands/bawwg/case/fl 1 .html).
1994
1995
1996 TARGETED DESIGN
1997
1998 A targeted approach to sampling design may be more appropriate when resources are limited.
1999 Targeted sampling is a specialized case of random stratified sampling. The approach described
2000 here involves defining a gradient of impairment. Once the gradient has been defined and systems
2001 have been placed in categories of impairment, investigators focus the greatest efforts on
2002 identifying and characterizing wetland systems or sites likely to be impacted by anthropogenic
2003 stressors, and on relatively undisturbed wetland systems or sites (see Identifying and
2004 Characterizing Reference Systems, Chapter 3), that can serve as regional, sub-regional, or
2005 watershed examples of natural biological integrity. Florida Department of Environmental
2006 Protection (FL DEP) uses a targeted sampling design for developing thresholds of impairment
2007 with macroinvertebrates (http://www.epa.gov/owow/wetlands/bawwg/case/fl2.html). Choosing
2008 sampling stations that best allow comparison of ecological integrity at reference wetland sites of
2009 known condition can conserve financial resources. A sampling design that tests specific
2010 hypotheses (e.g., the FL DEP study tested the effect of elevated water column phosphorus on
2011 macroinvertebrate species richness) generally can be analyzed with statistical rigor and can
2012 conserve resources by answering specific questions. Furthermore, identification of systems with
2013 problems and reference conditions eliminates the need for selecting a random sample of the
2014 population for monitoring.
2015
58
-------
December 2006- DRAFT Chapter 4. Sampling Design of Wetland Monitoring
2016 Targeted sampling assumes some knowledge of the systems sampled. Systems based on
2017 independent variables with evidence of degradation are compared to reference systems that are
2018 similar in their physical structure (i.e., in the same class of wetlands). For targeted sampling,
2019 wetlands should be characterized by a degree of impairment. Wetland systems should be viewed
2020 along a continuum from reference to degraded. An impaired or degraded wetland is a system in
2021 which anthropogenic impacts exceed acceptable levels, or interfere with beneficial uses.
2022 Comparison of the monitoring data to data collected from reference wetlands will allow
2023 characterization of the sampled systems. Wetlands identified as "at risk" should be evaluated
2024 through a sampling program to characterize the degree of degradation. Once characterized, the
2025 wetlands should be placed in one of the following categories:
i9i<7
2028 1. Degraded wetlands -wetlands in which the level of anthropogenic perturbance interferes
2029 with designated uses.
2030
2031 2. High-risk wetlands -wetlands where anthropogenic stress is high but does not
2032 significantly impair designated uses. In high-risk systems impairment is prevented by one
2033 or a few factors that could be changed by human actions, though characteristics of
2034 ecological integrity are already marginal.
2035
2036 3. Low-risk wetlands -wetlands where many factors prevent impairment, stressors are
2037 maintained below problem levels, and/or no development is contemplated that would
2038 change these conditions.
2039
2040 4. Reference wetlands -wetlands where the ecological characteristics most closely represent
2041 the pristine or minimally impaired condition.
2042
2043 Once wetland systems have been classified based on their physical structure (see chapter 3) and
2044 placed into the above categories, specific wetlands need to be selected for monitoring. At this
2045 point, randomness is introduced; wetlands should be randomly selected within each class and
2046 risk category for monitoring. An excellent example of categorizing wetlands in this manner is
2047 given in the Ohio Environmental Protection Agency's (OH EPA) case study [APPENDIX B],
2048 also available at: http://www.epa.gov/owow/wetlands/bawwg/case/oh 1 .html. They used the Ohio
2049 Rapid Assessment Method to categorize wetlands by degree of impairment. The Minnesota
2050 Pollution Control Agency (MPCA) also used a targeted design for monitoring wetlands
2051 (http://www.epa.gov/owow/wetlands/bawwg/case/mnl.html). They used the best professional
2052 judgment of local resource managers to identify reference sites and those with known
2053 impairment from identified stressors (agriculture and stormwater runoff).
2054
2055 Targeted sampling design involves monitoring identified degraded systems and comparable
2056 reference systems most intensively. Low risk systems are monitored less frequently (after initial
2057 identification), unless changes in the watershed indicate an increased risk of degradation.
2058
59
-------
December 2006- DRAFT Chapter 4. Sampling Design for Wetland Monitoring
2059 Activities surrounding impaired wetland systems may be used to help identify which actions
2060 negatively affect wetlands, and therefore may initiate more intensive monitoring of at-risk
2061 wetlands. Monitoring should focus on factors likely to identify ecological degradation and
2062 anthropogenic stress and on any actions that might alter those factors. State/Tribal water quality
2063 agencies should encourage adoption of local watershed protection plans to minimize ecological
2064 degradation of natural wetland systems. Development plans in the watershed should be evaluated
2065 to identify potential future stressors. Ecological degradation often gradually increases due to
2066 many growing sources of anthropogenic stress. Hence, frequent monitoring may be warranted
2067 for high-risk wetlands if sufficient resources remain after meeting the needs of degraded
2068 wetlands. Whenever development plans appear likely to alter factors that maintain ecological
2069 integrity in a high-risk wetland (e.g., vegetated buffer zones), monitoring should be initiated at a
2070 higher sampling frequency in order to enhance the understanding of baseline conditions (USEPA
2071 2000).
2072
2073 BEFORE/AFTER, CONTROL/IMPACT (BACI) DESIGN
2074
2075 An ideal before/after impact survey has several features: 1) the type of impact, time of impact,
2076 and place of occurrence should be known in advance; 2) the impact should not have occurred
2077 yet; and 3) control areas should be available (Green 1979). The first feature allows the surveys to
2078 be efficiently planned to account for the probable change in the environment. The second feature
2079 allows a baseline study to be established and to be extended as needed. The last feature allows
2080 the surveyor to distinguish between temporal effects unrelated to the impact and changes related
2081 to the impact. In practice however, advance knowledge of specific impacts is rare, and the ideal
2082 impact survey is rarely conducted. BACI designs modified to monitor impacts during or after
2083 their occurrence still can provide information, but there is an increase in the uncertainty
2084 associated with the results, and the likelihood of finding a statistically significant change due to
2085 the impact is less probable. In addition, other aspects of survey design are dependent on the
2086 study objectives, e.g., the sampling interval, the length of time the survey is conducted (i.e.,
2087 sampling for acute versus chronic effects), and the statistical analyses appropriate for analyzing
2088 the data (Suter 1993).
2089
2090 The best interval for sampling is determined by the objectives of the study (Kentula et al. 1993).
2091 If the objective is to detect changes in trends (e.g., regular monitoring for detection of changes in
2092 water quality or biotic integrity), regularly spaced intervals are preferred because the analysis is
2093 easier. On the other hand, if the objective is to assess differences before and after impact, then
2094 samples at random time points are advantageous. Random sample intervals reduce the likelihood
2095 that cyclic differences unforeseen by the sampler will influence the size of the difference before
2096 and after the impact. For example, surveys taken every summer for a number of years before and
2097 after a clear-cut may show little difference in system quality; however, differences may exist that
2098 can only be detected in the winter and therefore may go undetected if sampling occurs only
2099 during summer.
2100
60
-------
December 2006- DRAFT Chapter 4. Sampling Design for Wetland Monitoring
2101 The simplest impact survey design involves taking a single survey before and after the impact
2102 event (Green 1979). This type of design has the obvious pitfall that there may be no relationship
2103 between the observed event and the changes in the response variable- the change may be
2104 entirely coincidental. This pitfall is addressed in BACI design by comparing before and after
2105 impact data to data collected from a similar control system nearby. Data are collected before and
2106 after a potential disturbance in two areas (treatment and a control) with measurements on
2107 biological and environmental variables in all combinations of time and area (Green 1979). We
2108 will use a clear-cut adjacent to a wetland as an example to illustrate the BACI design. The
2109 sampling design is developed to identify the effects of clear-cutting on adjacent wetland systems.
2110 In the simplest BACI design, two wetlands would be sampled. One wetland would be adjacent to
2111 the clear-cut (the treatment wetland); the second wetland would be adjacent to a control site that
2112 is not clear-cut. The control site should have characteristics (soil, vegetation, structure,
2113 functions) similar to the treatment wetland, and is exposed to climate and weather similar to the
2114 first wetland. Both wetlands are sampled at the same time points before the clear-cut occurs and
2115 at the same time point after the clear-cut takes place. This design is technically known as an
2116 area-by-time factorial design. Evidence of an impact is found by comparing the control site
2117 samples (before and after) with the treatment site before and after samples. Area-by-time
2118 factorial design allows for both natural wetland-to-wetland variation and coincidental time
2119 effects. If there is no effect of the clear-cut, then change in system quality between the two time
2120 points should be the same. If there is an effect of the clear-cut, the change in system quality
2121 between the two time points should be different.
2122
2123 CONSIDERATIONS FOR BACI DESIGN
2124
2125 There are some potential problems with BACI design. First, because the control and impact sites
2126 are not randomly assigned, observed differences between sites may be related solely to some
2127 other factor that differs between the two sites. One could argue that it is unfair to ascribe the
2128 effect to the impact (Hurlbert 1984; Underwood 1991). However, as pointed out by Stewart-
2129 Oaten et al. (1986), the survey is concerned about a particular impact in a particular place, not in
2130 the average of the impact when replicated in many different locations. Consequently, it may be
2131 possible to detect a difference between these two specific sites. However, if there are no
2132 randomized replicate treatments, the results of the study cannot be generalized to similar events
2133 at different wetlands. However, the likelihood that the differences between sites are due to
2134 factors other than the impact can be reduced by monitoring several control sites (Underwood
2135 1991) because multiple control sites provide some information about potential effects of other
2136 factors.
2137
2138 The second and more serious concern with the simple Before-After design with a single
2139 sampling point before and after the impact, is that it fails to recognize that there may be natural
2140 fluctuations in the characteristic of interest that are unrelated to any impact (Hurlbert 1984;
2141 Stewart-Oaten 1986). Single samples before and after impact would be sufficient to detect the
2142 effects of the impact, if there were no natural fluctuations over time. However, if the population
61
-------
December 2006- DRAFT Chapter 4. Sampling Design for Wetland Monitoring
2143 also has natural fluctuations over and above the long-term average, then it is impossible to
2144 distinguish between cases where there is no effect from cases where there is an impact.
2145 Consequently, measured differences in system quality may be artifacts of the sampling dates and
2146 natural fluctuations may obscure differences or lead one to believe differences are present when
2147 they are not.
2148
2149 The simple BACI design was extended by Stewart-Oaten et al. (1986) by pairing surveys at
2150 several selected time points before and after the impact to help resolve the issue of
2151 psuedoreplication (Hulbert 1984). This modification of the BACI design is referred to as BACI-
2152 PS (Before-After, Control-Impact Paired Series design). The selected sites are measured at the
2153 same time points. The rationale behind this paired design is that repeated sampling before the
2154 impact gives an indication of the pattern of differences of potential change between the two sites.
2155 BACI-PS study design provides information both on the mean difference in the wetland system
2156 quality before and after impact, and on the natural variability of the system quality
2157 measurements. The resource manager has detected an effect if the changes in the mean
2158 difference are large relative to natural variability. Considerations for sampling at either random
2159 or regularly spaced intervals also apply here. Replication of samples should also be included if
2160 resources allow to improve certainty of analytical results.
2161
2162 Violation of the BACI assumptions may invalidate conclusions drawn from the data. Enough
2163 data should be collected before the impact to identify the trends in the communities of each
2164 sampling site if the BACI assumptions are to be met. Clearly defining the objectives of the study
2165 and identifying a statistically testable model of the relationships the investigator is studying can
2166 help resolve these issues (Suter 1993).
2167
2168 The designs described above are suitable for detecting longer-term chronic effects in the mean
2169 level of the variable of interest. However, the impact may have an acute effect (i.e., effects only
2170 last for a short while), or may change the variability in response (e.g., seasonal changes become
2171 more pronounced) in some cases. The sampling schedule can be modified so that it occurs at two
2172 temporal scales (enhanced BACI -PS design) that encompass both acute and chronic effects
2173 (Underwood 1991). The modified temporal design introduces randomization by randomly
2174 choosing sampling occasions in two periods (Before and After) in the control or impacted sites.
2175 The two temporal scales (sampling periods vs. sampling occasions) allow the detection of a
2176 change in mean and of a change in variability after impact. For example, groups of surveys could
2177 be conducted every year with five
2178 surveys one week apart randomly located within each group. The analysis of such a design is
2179 presented in Underwood (1991). Again, multiple control sites should be used to counter the
2180 argument that detected differences are specific to the sampled site. The September 2000 issue of
2181 the Journal of Agricultural, Biological, and Environmental Statistics discusses many of the
2182 advantages and disadvantages of the BACI design, and provides several examples of appropriate
2183 statistical analyses for evaluation of BACI studies.
2184
62
-------
December 2006- DRAFT Chapter 4. Sampling Design for Wetland Monitoring
2185
2186 4.4 SUMMARY
2187
2188 State and Tribal monitoring programs should be designed to assess wetland condition with
2189 statistical rigor while maximizing available management resources. The three approaches
2190 described in this module, probabilistic sampling, targeted/tiered approach, and BACI
2191 (Before/After, Control/Impact), present study designs that allow one to obtain a significant
2192 amount of information for statistical analyses. The sampling design selected for a monitoring
2193 program should depend on the management question being asked. Sampling efforts should be
2194 designed to collect information that will answer management questions in a way that will allow
2195 robust statistical analysis. In addition, site selection, characterization of reference sites or
2196 systems, and identification of appropriate index periods are all of particular concern when
2197 selecting an appropriate sampling design. Careful selection of sampling design will allow the
2198 best use of financial resources and will result in the collection of high quality data for evaluation
2199 of the wetland resources of a State or Tribe. Examples of different sampling designs currently in
2200 use for State and Tribal wetland monitoring are described in the Case Study module #14 on the
2201 website:
2202 http ://www. epa. gov/waterscience/criteria/wetlands/.
63
-------
December 2006- DRAFT
Chapter 4. Sampling Design for Wetland Monitoring
Table 4. Comparison of Probabilistic, Targeted, and BACI Sampling Designs
Probabilistic
Targeted
BACI
Random selection of wetland
systems from entire population
within a region.
This design requires minimal
prior knowledge of wetlands
within the sample population
for stratification.
This design may use more
resources (time and money) to
randomly sample wetland
classes, because more wetlands
may need to be sampled.
System characterization for a
class of wetlands is more
statistically robust.
Rare wetlands may be under-
represented or absent from the
sampled wetlands.
This design is potentially best
for regional characterization of
wetland classes, especially
water quality conditions are
not known.
Targeted selection of wetlands
based on problematic (wetland
systems known to have
problems) and reference
wetlands.
This design requires there to
be prior knowledge of
wetlands within the sample
population.
This design utilizes fewer
resources because only
targeted systems are sampled.
System characterization for a
class of wetlands is less
statistically robust, although
characterization of a targeted
wetland may be statistically
robust.
This design may miss
important wetland systems if
they are not selected for the
targeted investigation.
This design is potentially best
for site-specific and
watershed-specific criteria
development when water
quality conditions for the
wetland of interest are known.
Selection of wetlands based on
a known impact.
This design requires
knowledge of a specific
impact to be analyzed.
This design may use fewer
resources because only
wetlands with known impacts
and associated control systems
are sampled.
Characterization of the
investigated systems is
statistically robust.
The information gained in this
type of investigation is not
transferable to wetland
systems not included in the
study.
This design is potentially best
for monitoring restoration or
creation of wetlands and
systems that have specific
known stressors.
64
-------
December 2006- DRAFT Chapter 5. Candidate Variables for Establishing Nutrient Criteria
2203 Chapter 5 Candidate Variables for Establishing Nutrient
2204 Criteria
2205
2206 5.1 OVERVIEW OF CANDIDATE VARIABLES
2207
2208 This chapter provides an overview of candidate variables that could be used to establish nutrient
2209 criteria for wetlands. A good place to start with selecting candidate variables is by developing a
2210 conceptual model of how human activities affect nutrients and wetlands. These conceptual
2211 models may vary from complex to very simple models, such as relating nitrogen concentrations
2212 in sediments and plant biomass or species composition. Conceptual models establish the detail
2213 and scope of the project and the most important variables to select. In addition, they define the
2214 cause-effect relationships that should be documented to determine whether a problem occurs and
2215 what is causing the problem.
2216
2217 In general, for the purposes of numeric nutrient criteria development, it is helpful to develop an
2218 understanding of the relationships among human activities, nutrients and habitat alterations, and
2219 attributes of ecosystem structure and function, to establish a simple causal pathway among three
2220 basic elements in a conceptual model. These three basic groups of variables are important to
2221 distinguish because we use them differently in environmental management (Stevenson et al.
2222 2004a). A fourth group of variables is important in order to account for variation in expected
2223 condition of wetlands due to natural variation in landscape setting.
2224
2225 The overview of candidate variables in this chapter follows the outline provided in the
2226 conceptual model in Figure 5.1. Historically, variables in conceptual models have been grouped
2227 many ways with a variety of group names (Paulsen et al. 1991; USEPA 1996; 1998a; Stevenson
2228 1998; Stevenson 2004a, b). In this document, three groups and group names are used to
2229 emphasize cause-effect relationships, simplify their presentation and discussion for a diversity of
2230 audiences, and maintain some continuity between their use in the past and their use here. The
2231 three groups are supporting variables, causal variables, and response variables.
2232
2233 Supporting variables provide information useful in normalizing causal and response variables
2234 and categorizing wetlands. (These are in addition to characteristics used to define wetland
2235 classes as described in Chapter 3 ) Causal variables characterize pollution or habitat
2236 alterations. Causal variables are intended to characterize nutrient availability in wetlands and
2237 could include nutrient loading rates and soil nutrient concentrations. Response variables are
2238 direct measures or indicators of ecological properties. Response variables are intended to
2239 characterize biotic response and could include community structure and composition of
2240 vegetation and algae. The actual grouping of variables is much less important than
2241 understanding relationships among variables.
2242
65
-------
December 2006- DRAFT Chapter 5. Candidate Variables for Establishing Nutrient Criteria
2243 It is important to recognize the complex temporal and spatial structure of wetlands when
2244 measuring or interpreting causal and response variables with respect to nutrient condition. The
2245 complex interaction of climate, geomorphology, soils and internal interactions has led to a
2246 diverse array of wetland types, ranging from infrequently flooded, isolated depress!onal wetlands
2247 such as seasonal prairie potholes and playa lakes to very large complex systems such as the
2248 Everglades and the Okefenokee Swamp. In addition, most wetlands are complex temporal and
2249 spatial mosaics of habitats with distinct structural and functional characteristics, illustrated most
2250 visibly by patterns in vegetation structure.
2251
2252 Horizontal zonation is a common feature of wetland ecosystems, and in most wetlands, relatively
2253 distinct bands of vegetation develop in relation to water depth. Bottomland hardwood forests and
2254 prairie pothole wetlands provide excellent illustrations of zonation in two very divergent wetland
2255 types. However, vegetation zones are not static. Seasonal and long-term changes in vegetation
2256 structure are a common characteristic of most wetland ecosystems. Wetlands may exhibit
2257 dramatic shifts in vegetation patterns in response to changes in hydrology, with entire wetlands
2258 shifting between predominantly emergent vegetation to completely open water within only a
2259 year or two. Such temporal patterns in fact are important features of many wetlands and should
2260 be considered in interpreting any causal or response variable. For example, seasonal cycles are
2261 an essential feature of floodplain forests, which are typically flooded during high spring flows
2262 but dry by mid to late summer. Longer-term cycles are similarly essential features of prairie
2263 pothole wetlands, which exhibit striking shifts in vegetation in response to water level
2264 fluctuations over periods of a few years in smaller wetlands to decades in larger, more permanent
2265 wetlands (van der Valk 2000). Vegetation patterns are likely to control major aspects of wetland
2266 biogeochemistry, and trophic dynamics can significantly affect the physical and chemical
2267 characteristics of sediments and overlying waters (Rose and Crumpton 1996).
2268
2269 The complex temporal and spatial structure of wetlands should influence the selection of
2270 variables to measure and methods for measuring them. Most wetlands are characterized by
2271 extremely variable hydrologic and nutrient loading rates and close coupling of soil and water
2272 column processes. As a result, estimates of nutrient loading may prove more useful than direct
2273 measurements of water column nutrient concentrations as causal variables for establishing the
2274 nutrient condition of wetlands. In addition, soil nutrients that integrate a wetland's variable
2275 nutrient history over a period of years may provide the most useful metric against which to
2276 evaluate wetland response.
66
-------
December 2006- DRAFT
Chapter 5. Candidate Variables for Establishing Nutrient Criteria
Fig 5.1
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
Natural Factors
Climate
Geology
Topography
T
Human Activities
Agricult
»
Ripar
jre Urban Dev. C
\s<
an Buffers
No-till Rov
Channelization
i/ Crops
T
Stressors: Contaminants & Habitat Alterations
Nitrogen
X
Phosphorus
/ '
Invasive Species
X
Hydrologic
Alterations
This conceptual model illustrates
the causal pathway between
human activities and valued
ecological attributes. It includes
the role of nutrients in a broader
context that includes natural
variation among wetlands. The
relationship between different
approaches of grouping variables
is illustrated to emphasize the
importance of cause-effect
relationships. Here, natural factors
and human activities regulate the
physical, chemical and biological
attributes of wetlands. Some
wetland attributes are more valued
than others and provide the
endpoints of assessment and
management. Some physical,
chemical, and biological attributes
are stressors, i.e. contaminants
and habitat alterations caused by
human activities that negatively
affect valued ecological attributes.
The overview of variables in
Chapter 5 is organized in three
sections: supporting, causal, and
response variables. Supporting
variables are natural landscape-
level factors that classify expected
condition of wetlands. Causal
factors "cause" effects in response
variables.
5.2 SUPPORTING VARIABLES
Supporting variables are not intended to characterize nutrient availability or biotic response but
rather to provide information that can be useful in normalizing causal and response variables.
Below is a brief overview of supporting variables that might be useful for categorizing wetlands
and for normalizing and interpreting causal and response variables. Please refer to EPA module
#18 Biogeochemical Indicators for a more detailed description of soil variables and to EPA
module #21 Wetland Hydrology for a more detailed description of hydrologic condition.
CONDUCTIVITY
Conductivity (also called electrical conductance or specific conductance) is an indirect measure
of total dissolved solids. This is due to the ability of water to conduct an electrical current when
there are dissolved ions in solution - water with higher concentrations of dissolved inorganic
compounds have higher conductivity. Conductivity is commonly measured in situ using a
handheld probe and conductivity meter (APHA 1999), or using automated conductivity loggers.
Responses: Valued Ecological Attributes
Ecosystem Structure
•Biomass
•Biodiversity
Flora & Fauna
Ecosystem Function
•Productivity
•Nutrient Retention
•Hydrologic Regulation
67
-------
December 2006- DRAFT Chapter 5. Candidate Variables for Establishing Nutrient Criteria
2295 Because the conductivity changes with temperature, the raw measurement should be adjusted to
2296 a reference temperature of 25°C. A multiplier of 0.7 is commonly applied to estimate the total
2297 dissolved solids concentration (mg/L) in fresh water when the conductivity is measured in units
2298 of microSiemens per centimeter (jiS/cm), although this multiplier varies with the types of
2299 dissolved ions and should be adjusted for local chemical conditions.
2300
2301 Conductivity is a useful tool for characterizing wetland inputs and interpreting nutrient condition
2302 because of its sensitivity to changes in these inputs. Rainfall tends to have lower conductivity
2303 than surface water, with ground water often having higher values due to the longer residence
2304 time of water in the subsurface. Coastal and marine waters - as well as water in terminal lakes
2305 and wetlands - have even higher conductivity due to the influence of salinity. Municipal and
2306 industrial discharges often have higher conductivity than their intake waters due to the addition
2307 of soluble wastes. Wetland hydrologic inputs can be identified by comparing the measured input
2308 conductivity with the conductivity of potential local sources.
2309
2310 SOILPH
2311
2312 Soil pH can be important for categorizing wetland soils and interpreting soil nutrient variables.
2313 The pH of wetland soils and water varies over a wide range of values. Many ombrotrophic
2314 organic wetland soils (histosols) such as bogs, non-limestone based wetlands are often acidic and
2315 mineral wetland soils are frequently neutral or alkaline. Flooding a soil results in consumption of
2316 electrons and protons. In general, flooding acidic soils results in an increase in pH, and flooding
2317 alkaline soils decreases pH (Mitsch and Gosselink, 1993). The increase in pH of low pH (acidic)
2318 wetland soils is largely due to the reduction of iron and manganese oxides. However, the initial
2319 decrease in pH of alkaline wetland soils is due to rapid decomposition of soil organic matter and
2320 accumulation of CO2. The decrease in pH that generally occurs when alkaline soils are flooded
2321 results from the buildup of CC>2 and carbonic acid. In addition, the pH of alkaline soils is highly
2322 sensitive to changes in the partial pressure of CC>2. Carbonates of iron and manganese also can
2323 buffer the pH of soil to neutrality. Soil pH determinations should be made on wet soil samples.
2324 Once the soils are air-dried, oxidation of various reduced compounds results in decrease in pH
2325 and the values may not represent ambient conditions.
2326
2327 Soil pH is measured using commercially available combination electrodes on soil slurries. If air
2328 dry or moist soil is used, a 1:1 soil to water ratio should be used. For details on methodology, the
2329 reader is referred to Thomas (1996).
2330
2331 Soil pH can explain the availability and retention capacity of phosphorus. For example,
2332 phosphorus bioavailability is highest at soil pH near neutral conditions. For mineral soils,
2333 phosphorus adsorption capacity has been directly linked to extractable iron and aluminum. For
2334 details the reader is referred to Module-18 on Biogeochemical Indicators.
2335
68
-------
December 2006- DRAFT Chapter 5. Candidate Variables for Establishing Nutrient Criteria
2336 SOIL BULK DENSITY
2337
2338 Soil bulk density is the mass of dry solids per unit volume of soil, which includes the volume of
2339 solids plus air- and water-filled pore space. Bulk density is a useful parameter for expressing the
2340 concentration of nutrients on a volume basis rather than mass basis. For example, concentration
2341 of nutrients in organic wetland soils can be high when expressed on a mass basis (mg/kg or (ig/g
2342 of dry soil), as compared to mineral wetland soils. However, the difference in concentration may
2343 not be as high when expressed on a volume (cm3) basis, which is calculated as the product of
2344 bulk density and nutrient concentration per gram of soil. Expressing soil nutrient concentrations
2345 on a volume basis is especially relevant to uptake by vegetation since plant roots explore a
2346 specific volume, not mass, of soil. Expressing nutrients on a volume basis also helps in
2347 calculating total nutrient storage in a defined soil layer.
2348
2349 Bulk density is measured by collecting an intact soil core of known volume at specific depths in
2350 the soil (Blake and Hartge, 1986). Cores are oven-dried at 70°C and weighed. Bulk density is
2351 calculated as follows:
2352
2353 Bulk density (dry) (g/cm3) = mass dry weight (grams)/volume (cm3)
2354
2355 Bulk densities of wetland organic soils range from 0.1 to 0.5 g/cm3, whereas bulk densities of
2356 mineral wetland soils range from 0.5 to 1.5 g/cm3. Soil bulk densities are directly related to soil
2357 organic matter content, as bulk densities decrease with increase in soil organic matter content.
2358
2359 SOIL ORGANIC MATTER CONTENT
2360
2361 Soil organic matter can be important for categorizing wetland soils and interpreting soil nutrient
2362 variables. Wetland soils often are characterized by the accumulation of organic matter because
2363 rates of primary production often exceed rates of decomposition. Some wetlands accumulate
2364 thick layers of organic matter that, over time, form peat soil. Organic matter provides nutrient
2365 storage and supply, increases the cation exchange capacity of soils, enhances adsorption or
2366 deactivation of organic chemicals and trace metals, and improves overall soil structure, which
2367 results in improved air and water movement. A number of methods are now routinely used to
2368 estimate soil organic matter content expressed as total organic carbon or loss on ignition (APHA,
2369 1999; Nelson and Sommers, 1996).
2370
2371 Soil organic matter content represents the soil organic carbon content of soils. Typically, soil
2372 organic matter content is approximately 1.7 to 1.8 times that of total organic carbon. The carbon
2373 to nitrogen and carbon to phosphorus ratios of soils can provide an indication of nutrient
2374 availability in soils.
69
-------
December 2006- DRAFT Chapter 5. Candidate Variables for Establishing Nutrient Criteria
2375
2376 HYDROLOGIC CONDITION
2377
2378 Wetland hydrologic condition is important for characterizing wetlands and for normalizing many
2379 causal and response variables. Hydrologic conditions can directly affect the chemical and
2380 physical processes governing nutrient and suspended solids dynamics within wetlands (Mitsch
2381 and Gosselink, 2000). Detailed, site-specific hydrologic information available is best, but at a
2382 minimum, some estimate of water level fluctuation should be made. A defining characteristic of
2383 wetlands is oxygen deficiency in the soil caused by flooding or soil saturation. These conditions
2384 influence vegetation dynamics through differential growth and survival of plant species and also
2385 exert significant control over biogeochemical processes involved in carbon flow and nutrient
2386 cycling within wetlands. Spatial and temporal patterns in hydrology can create complex patterns
2387 in soil and water column oxygen availability including alternating aerobic and anaerobic
2388 conditions in wetland soils, with obvious implications for plant response and biogeochemical
2389 process dynamics. Water levels in wetlands can be determined using a staff gauge when surface
2390 water is present. A staff gauge measures the depth of surface flooding relative to a reference
2391 point such as the soil surface. While surface flooding may be rare or absent in many wetlands,
2392 high water tables may still cause soil saturation in the rooting zone. In wetlands where soils are
2393 saturated, water level can be measured with a small diameter perforated tube installed in the soil
2394 to a specified depth (Amoozegar and Warrick 1986). Automated water level recorders using
2395 floats, capacitance probes, or pressure transducers are suitable for measuring water levels both
2396 above- and below-ground.
2397
2398
2399 5.3 CAUSAL VARIABLES
2400 Causal variables are intended to characterize nutrient availability in wetlands. Most wetlands are
2401 characterized by extremely variable nutrient loading rates and close coupling of soil and water
2402 column processes. As a result, estimates of nutrient loading and measurements of soil nutrients
2403 may prove more useful than direct measurements of water column nutrient concentrations as
2404 causal variables for establishing the nutrient condition of wetlands. Nutrient loading history and
2405 soil nutrient measures can integrate a wetland's variable nutrient history over a period of years
2406 and may provide especially useful metrics against which to evaluate nutrient condition. Wetlands
2407 exhibit a high degree of spatial heterogeneity in chemical composition of soil layers, and areas
2408 impacted by nutrients may exhibit more variability than unimpacted areas of the same wetland.
2409 Thus, sampling protocols should capture this spatial variability. Developing nutrient criteria and
2410 monitoring the success of nutrient management programs involves important considerations for
2411 sampling designed to capture spatial and temporal patterns. (See Study Design Module and the
2412 Biogeochemical Indicators Module.)
2413
2414 Below is a brief overview of the use of nutrient loading and soil and water column nutrient
2415 measures for estimating nutrient condition of wetlands. Please refer to the EPA module #19
2416 Nutrient Loading Models for a detailed description of nutrient load estimation and to EPA
70
-------
December 2006- DRAFT Chapter 5. Candidate Variables for Establishing Nutrient Criteria
2417 module #18 Biogeochemical Indicators for a more detailed description of soil and water column
2418 nutrient measures.
2419
2420 NUTRIENT LOADING
2421
2422 External nutrient loads to wetlands are determined primarily by surface and subsurface transport
2423 from the contributing landscape, and vary significantly as a function of weather and landscape
2424 characteristics such as soils, topography, and land use. Most wetlands are characterized by
2425 extremely variable hydrologic and nutrient loading rates, which present considerable obstacles to
2426 obtaining adequate direct measurement of nutrient inputs. Adequate measurement of loads may
2427 require automated samplers capable of providing flow-weighted samples when loading rates are
2428 highly variable. In many cases, non-point source loads simply may not be adequately sampled.
2429 The more detailed the loading measurements the better, but it is not reasonable to expect
2430 adequate direct measurement of loads for most wetlands. In the absence of sufficient, direct
2431 measurements, it may be possible to estimate nutrient loading using an appropriate loading
2432 model or at least to provide a relative ranking of wetlands based on expected nutrient load. One
2433 advantage of loading models is that nutrient loading can be integrated over the appropriate time
2434 scale for characterizing wetland nutrient condition and, in some cases, historical loading patterns
2435 can be reconstructed. Loading models also can provide hydrologic loading rates to calculate
2436 critical supporting variables such as hydroperiod and residence times.
2437
2438 Loading function models are based on empirical or semi-empirical relationships that provide
2439 estimates of pollutant loads on the basis of long-term measurements of flow and contaminant
2440 concentration. Generally, loading function models contain procedures for estimating pollutant
2441 load based on empirical relationships between landscape physiographic characteristics and
2442 phenomena that control pollutant export. McElroy et al. (1976) and Mills (1985) described
2443 loading functions employed in screening models developed by the USEPA to facilitate
2444 estimation of nutrient loads from point and nonpoint sources. The models contain simple
2445 empirical expressions that relate the magnitude of nonpoint pollutant load to readily available or
2446 measurable input parameters such as soils, land use and land cover, land management practices,
2447 and topography. Preston and Brakebill (1999) described a spatial regression model that relates
2448 the water quality conditions within a watershed to sources of nutrients and to those factors that
2449 influence transport of the nutrients. The regression model, Spatially-Referenced Regressions on
2450 Watersheds (SPARROW) involves a statistical technique that utilizes spatially referenced
2451 information and data to provide estimates of nutrient load (Smith et al., 1997; Smith et al., 2003;
2452 http://water.usgs.gov/nawqa/sparrow/).
2453
2454 In general, the SPARROW methodology was designed to provide statistically based
2455 relationships between stream water quality and anthropogenic factors such as contaminant
2456 sources within the contributing watersheds, land surface characteristics that influence the
2457 delivery of pollutants to the stream, and in-stream contaminant losses via chemical and
2458 biological process pathways. The Generalized Watershed Loading Functions (GWLF) model
71
-------
December 2006- DRAFT Chapter 5. Candidate Variables for Establishing Nutrient Criteria
2459 (Haith and Shoemaker, 1987; Haith et al., 1992) uses daily time steps, and to some extent, both
2460 can be used to examine seasonal variability and the response to landscape characteristics of
2461 specific watersheds. The GWLF model was developed to evaluate the point and non-point
2462 loading of nitrogen and phosphorus in urban and rural watersheds. The model enhances
2463 assessment of effectiveness of certain land use management practices and makes extensive use of
2464 readily available watershed data. The GWLF also provides an analytical tool to identify and rank
2465 critical areas of a watershed and to evaluate alternative land management programs.
2466
2467 Process-oriented simulation models attempt to explicitly represent biological, chemical, and
2468 physical processes controlling hydrology and pollutant transport. These models are at least partly
2469 mechanistic in nature and are built from equations that contain directly definable, observable
2470 parameters. Examples of process-oriented simulation models that have been used to predict
2471 watershed hydrology and water quality include the Agricultural Nonpoint Source model
2472 (AGNPS), the Hydrologic Simulation Program-Fortran (HSPF), and the Soil and Water
2473 Assessment Tool (SWAT). AGNPS (Young et al.1987) is a distributed parameter, event-based
2474 and continuous simulation model that predicts the behavior of runoff, sediment, nutrients and
2475 pesticide transport from watersheds that have agriculture as the primary land use. Because of its
2476 simplicity and ease of use, AGNPS is probably one of the most widely used hydrologic and
2477 water quality models of watershed assessment. HSPF (Johansen et al., 1984; Bicknell et al.,
2478 1993; Donigian et al., 1995a) is a lumped parameter, continuous simulation model developed
2479 during the mid-1970's to predict watershed hydrology and water quality for both conventional
2480 and toxic organic pollutants. HSPF is one of the most comprehensive models available for
2481 simulating non-point source nutrient loading. The capability, strengths, and weaknesses of HSPF
2482 have been demonstrated by its application to many urban and rural watersheds and basins (e.g.,
2483 Donigian et al., 1990; Moore et al., 1992; and Ball et al., 1993). SWAT (Arnold et al., 1995) is a
2484 lumped parameter, continuous simulation model developed by the USDA-Agricultural Research
2485 Services that provides long-term simulation of impact of land management practices on water,
2486 sediment, and agricultural chemical yields in large complex watersheds. Because of its lumped
2487 parameter nature, coupled with its extensive climatic, soil, and management databases, the
2488 SWAT model is one of the most widely used hydrologic and water quality models for large
2489 watersheds and basins, and the model has found widespread application in many modeling
2490 studies that involve systemic evaluation of impact of agricultural management on water quality.
2491
2492 These loading models address only gross, external nutrient inputs. It is important to consider the
2493 overall mass balance for the receiving wetland in developing measures of nutrient loading
2494 against which to evaluate wetland nutrient condition. This requires some estimate of nutrient
2495 export, storage, and transformation. In the absence of sufficient, direct measurements from
2496 which to calculate nutrient mass balance, it may be possible to estimate nutrient mass balances
2497 using an appropriate wetland model. Strictly empirical, regression models can be used to
2498 estimate nutrient retention and export in wetlands but these regressions are of little value outside
2499 the data domain in which they are developed. When developed for a diverse set of systems, the
2500 scatter in these regressions can be quite large. In contrast to strictly empirical regressions, mass
72
-------
December 2006- DRAFT Chapter 5. Candidate Variables for Establishing Nutrient Criteria
2501 balance models incorporate principles of mass conservation. These models integrate external
2502 loading to the wetland, nutrient transformation and retention within the wetland, and nutrient
2503 export from the wetland. Mass balance models allow time varying hydrologic and nutrient inputs
2504 and can provide estimates of spatial nutrient distribution with the wetland. The most difficult
2505 problem is developing removal rate equations which adequately represent nutrient
2506 transformation and retention across the range of conditions for which estimates are needed.
2507
2508 LAND USE
2509
2510 Identifying land uses in regions surrounding wetlands are important for characterizing reference
2511 condition, identifying reference wetlands, and providing indicators of nutrient loading rates for
2512 criteria development. Most simply, the percentage of natural area, or the percentage of
2513 agricultural and urban lands can be used to characterize land uses around wetlands. More
2514 detailed quantitative data can be gathered from GIS analyses which provides higher resolution
2515 identification of land use types, such as pastures, row crops, and confined animal feeding
2516 operations for agriculture. Ideally these characterizations should be done for the entire
2517 sourceshed, including both air and water, in the regions around wetlands. Air-sheds should
2518 incorporate potential atmospheric sources of nutrients, and watersheds should incorporate
2519 potential aquatic sources. However, in practice, land use around wetlands is typically used for
2520 defining reference wetlands and is used in most nutrient loading models to characterize
2521 groundwater and surface water sources. Land use in buffer zones, one kilometer zones around
2522 wetlands and wetland watersheds (delineated by elevation) have been used to characterize
2523 human activities that could be affecting wetlands.
2524
2525 EXTRACTABLE SOIL NITROGEN AND PHOSPHORUS
2526
2527 Ammonium is the dominant form of inorganic N in wetland soils, and unlike total soil N (Craft
2528 et al. 1995, Chiang et al. 2000), soil extractable NH4-N increases in response to N loadings.
2529 Enrichment leads to enhanced cycling of N between wetland biota (Valiela and Teal 1974,
2530 Broome et al. 1975, Chalmers 1979, Shaver et al. 1998), greater activity of denitrifying bacteria
2531 (Johnston 1991, Groffman 1994, White and Reddy 1999) and accelerated organic matter and N
2532 accumulation in soil (Reddy et al. 1993, Craft and Richardson 1998). In most cases, extractable
2533 soil N should be measured in the surface soil where roots and biological activity are
2534 concentrated.
2535
2536 Extractable N is measured by extraction of inorganic (NH4-N) N with 2 M KC1 (Mulvaney
2537 1996). Ten to twenty grams of field moist soil is equilibrated with 100 ml of 2 M KC1 for one
2538 hour on a reciprocating shaker followed by filtration through Whatman No. 42 filter paper.
2539 Ammonium-N in soil extracts is determined colorimetrically using the phenate or salicylate
2540 method (APHA 1999, Method 350.2, USEPA, 1993a).
2541
73
-------
December 2006- DRAFT Chapter 5. Candidate Variables for Establishing Nutrient Criteria
2542 Extractable P often is a reliable indicator of the P enrichment of soils, and in wetlands,
2543 extractable P is strongly correlated with surface water P concentration and P enrichment from
2544 external sources (Reddy et al. 1995, 1998). Selected methods used to extract P are described
2545 below (Kuo 1996). Many soil testing laboratories perform these analyses on a routine basis.
2546 Historically, these methods have been used to determine nutrient needs of agronomic crops, but
2547 the methods have been used more recently to estimate P impacts in upland and wetland soils
2548 (Sharpley et al. 1992; Nair et al. 1995; Reddy et al. 1995; 1998).
2549
2550 The Mehlich I method is typically used in Southeast and Mid-Atlantic region on mineral soils
2551 with pH of < 7.0 (Kuo 1996). The extractant consists of dilute concentrations of strong acids.
2552 Many plant nutrients such as P, K, Ca, Mg, Fe, Zn, and Cu extracted with Mehlich I methods
2553 have been calibrated for production of crops in agricultural ecosystems. This solvent extracts
2554 some Fe and Al- bound P, and some Ca-bound P. Soil (dry) to extractant ratio is set at 1: 4, for
2555 mineral soils, while wider ratios are used for organic soils. Soil solutions are equilibrated for
2556 period of 5 minutes on a mechanical shaker and filtered through Whatman No. 42 filter. Filtered
2557 solutions are analyzed for P and other nutrients using standard methods (Method 365. 1, USEPA,
2558 1993a).
2559
2560 The Bray P-l method has been widely used as an index of available P in soils (Kuo 1996). The
2561 combination of dilute concentration of strong acid (HC1 at 0.025 M) and ammonium fluoride
2562 (NH4F at 0.03 M) is designed to remove easily acid extractable soluble P forms such as Ca-
2563 bound P, and some Fe and Al-bound P. Soil (dry) to extractant ratio is set at 1: 7 for mineral soils
2564 with wider ratios used for highly organic soils, then shaken for 5 minutes and filtered through
2565 Whatman No. 42 filter. Filtered solutions are analyzed for P and other nutrients using the same
2566 methods used for the Mehlich I extraction (Method 365. 1, USEPA 1993a).
2567
2568 Bicarbonate Extractable P is a suitable method for calcareous soils. Soil P is extracted from the
2569 soil with 0.5 M NaHCOs, at nearly a constant pH of 8.5 (Kuo 1996). In calcareous, alkaline, or
2570 neutral soils containing Ca-bound P, this extractant decreases the concentration of Ca in solution
2571 by causing precipitation of Ca as CaCOs; and as result P concentration in soil solution increases.
2572 Soil (dry) to extraction ratio is set at 1: 20 for mineral soils and 1:100 for highly organic soils.
2573 Soil solutions are equilibrated for period of 30 minutes on a shaker and filtered through
2574 Whatman No. 42 filter paper and analyzed for P using standard methods (Method 365. 1,
2575 USEPA, 1993a).
2576
2577 TOTAL SOIL NITROGEN AND PHOSPHORUS
2578
2579 Nutrient enrichment leads to enrichment of total soil P (Craft and Richardson 1993, Reddy et al.
2580 1993, Bridgham et al., 2001). In contrast, soil total N usually does not increase in response to
2581 nutrient enrichment (Craft et al. 1995, Chiang et al. 2000). Rather, enrichment leads to enhanced
2582 cycling of N between wetland biota that is reflected in greater N uptake and net primary
2583 production (NPP) of wetland vegetation (Valiela and Teal 1974, Broome et al. 1975, Chalmers
74
-------
December 2006- DRAFT Chapter 5. Candidate Variables for Establishing Nutrient Criteria
2584 1979, Shaver et al. 1998), greater activity of denitrifying bacteria (Johnston 1991, Groffman
2585 1994, White and Reddy 1999) and accelerated organic matter and N accumulation in soil (Reddy
2586 et al. 1993, Craft and Richardson 1998). In most cases, total N and P should be measured in at
2587 least the surface soil where most roots and biological activity are concentrated.
2588
2589 Since ammonium N is the dominant form of inorganic nitrogen in saturated wetland soils with
2590 very little nitrate (NO3) present, total Kjeldahl nitrogen (TKN) can generally be taken as a
2591 measure of total N in such soils. The difference between TKN and ammonium N provides
2592 information on soil organic N. The soil organic carbon to soil organic nitrogen ratio of soils can
2593 provide an indication soils capacity to mineralize organic N and provide ammonium N to
2594 vegetation. TKN in soils is determined by converting organic forms of N to NH4-N by digestion
2595 with concentrated H2SO4 at temperatures of 300-350 ° C (Bremner 1996). The NH4-N in digested
2596 samples is analyzed using colorimetric (e.g., phenate, salicylate) methods (APHA 1999,
2597 Mulvaney 1996).
2598
2599 Total P in soils is determined by oxidation of organic forms of P and acid (nitric-perchloric acid)
2600 dissolution of minerals at temperatures of <300°C (Kuo 1996). Digested solutions are analyzed
2601 for P using colorimetric methods (e.g., ascorbic acid-molybdate) (APHA 1999, Kuo 1996).
2602 Many laboratories may not have access to perchloric acid fume-hoods. Alternatively, soil total
2603 phosphorus can be determined using ashing method (Anderson, 1976). Results obtained from
2604 this method are reliable and comparable to total phosphorus measurements made using
2605 perchloric acid digestion method.
2606
2607 WATER COLUMN NITROGEN AND PHOSPHOROUS
2608
2609 Nutrient inputs to wetlands are highly variable across space and time, however, so that single
2610 measurements of water column N and P represent only a "snap-shot" of nutrient condition, and
2611 may or may not reflect the long-term pattern of nutrient inputs that alter biogeochemical cycles
2612 and affect wetland biota. The best use of water column N and P concentrations for nutrient
2613 criteria development will be based on frequent monitoring of nutrient concentrations over time
2614 (e.g., weekly or monthly measurements). Of course, in wetlands that are seldom flooded,
2615 measurements of water column N and P may not be practical or even relevant for assessing
2616 impacts. Whenever, water samples are obtained, it is important the water depth is recorded, as
2617 nutrient concentration is related to water depth. In the case of tidal estuarine or freshwater
2618 wetlands, it is also important to record flow and the point in the tidal cycle that the samples were
2619 collected.
2620
2621 Methodologies to monitor N in surface waters are well developed for other ecosystems and can
2622 be readily adopted for wetlands. The most commonly monitored N species are total Kjeldahl
2623 nitrogen (TKN), ammonium N, and nitrate plus nitrite N (APHA 1999). The TKN analysis
2624 includes both organic and ammonium N, but does not include nitrate plus nitrite N. Organic N is
2625 determined as the difference between TKN and NH4-N. Forms of N in surface water are
75
-------
December 2006- DRAFT Chapter 5. Candidate Variables for Establishing Nutrient Criteria
2626 measured by standard methods, including phenol-hypochlorite for ammonium N, cadmium
2627 reduction of nitrate to nitrite for nitrate N and Kjeldahl digestion of total N to ammonium for
2628 analysis of total N (APHA 1999). Dissolved organic N is primarily used by heterotrophic
2629 microbes whereas plants and various microorganisms take up inorganic forms of N (ammonium
2630 N and nitrate N) to support metabolism and new growth.
2631
2632 Methodologies to monitor P in surface waters are well developed for aquatic ecosystems and can
2633 be readily adopted for wetlands (APHA 1999). The most commonly measured forms of P in
2634 surface water are total P, dissolved inorganic P (i.e., PO/t-P), and total dissolved P. To trace the
2635 transport and transformations of P in wetlands, it might be useful to distinguish four forms of P:
2636 (i) dissolved inorganic P (DIP, also referred to as dissolved reactive P (DRP) or soluble reactive
2637 phosphorous (SRP)); (ii) dissolved organic P (DOP); (iii) particulate inorganic P (PIP), and (iv)
2638 particulate organic P (POP). Dissolved inorganic P (PCVP) is considered bioavailable (e.g.,
2639 available for uptake and use by microorganisms, algae and vegetation) whereas organic and
2640 particulate P forms generally must be transformed into inorganic forms before being considered
2641 bioavailable. In P limited wetlands, a significant fraction of DOP can be hydrolyzed by
2642 phosphatases and utilized by both bacteria, algae, and macrophytes.
2643
2644
2645 5.4 RESPONSE VARIABLES
2646
2647 Biotic measures that can integrate a wetland's variable nutrient history over a period of months
2648 to years may provide the most useful measures of wetland response to nutrient enrichment.
2649 Microorganisms, algae and macrophytes respond to nutrient enrichment by (1) increasing the
2650 concentration of nutrients (P, N) in their tissues, (2) increasing growth and biomass production
2651 and (3) shifts in species composition. The biotic response to nutrient enrichment generally occurs
2652 in a sequential manner as nutrient uptake occurs first, followed by increased biomass production
2653 followed by a shift in species composition as some species disappear and other species replace
2654 them. Macroinvertebrates respond to nutrient enrichment indirectly as a result of changes in food
2655 sources, habitat structure, and dissolved oxygen. Because of their short life cycle,
2656 microorganisms and algae respond more quickly to nutrient enrichment than macrophytes.
2657 However, biotic measures that can integrate a wetland's variable nutrient history over a period of
2658 months to years may provide the most useful measures of wetland response.
2659
2660 Below is a brief overview of the use of macrophytes, algae, and macroinvertebrates to assess
2661 nutrient condition of wetlands. Please refer to the relevant modules in the EPA series "Methods
2662 for Evaluating Wetland Condition" for details on using vegetation
2663 (http://www.epa.gov/waterscience/criteria/wetlands/16Indicators.pdf:
76
-------
December 2006- DRAFT Chapter 5. Candidate Variables for Establishing Nutrient Criteria
2664 http://www.epa.gov/waterscience/criteria/wetlands/10Vegetation.pdf), algae
2665 (http://www.epa.gov/waterscience/criteria/wetlands/llAlgae.pdf), and macroinvertebrates
2666 (http://www.epa.gov/waterscience/criteria/wetlands/9Invertebrate.pdf). to assess wetland
2667 condition., including nutrients.
2668
2669 MACROPHYTE NITROGEN AND PHOSPHORUS
2670
2671 Wetland macrophytes respond to nutrient enrichment by increasing uptake and storage of N and
2672 P (Verhoeven and Schmitz 1991, Shaver et al. 1998, Chiang et al. 2000). In wetlands where P is
2673 the primary limiting nutrient, the P content of vegetation increases almost immediately (within a
2674 few months) in response to nutrient enrichment (Craft et al. 1995). Increased P uptake by plants
2675 is known as "luxury uptake" because P is stored in vacuoles and used later (Davis 1991). Like P,
2676 leaf tissue N may increase in response to N enrichment (Brinson et al. 1984, Shaver et al. 1998).
2677 However, most N is directly used to support new plant growth so that luxury uptake of N is not
2678 usually observed (Verhoeven and Schmitz 1991). Tidal marsh grasses, however do appear to
2679 store nitrogen in both living and dead tissues that can be accessed by living plant tissue. A
2680 discussion of conservation and translocation of N in saltwater tidal marshes can be found in
2681 Hopkinson and Schubauer 1980, and in Thomas and Christian 2001.
2682
2683 Nutrient content of macrophyte tissue holds promise as a means to assess nutrient enrichment of
2684 wetlands. However, several caveats should be kept in mind when using this diagnostic tool
2685 (Gerloff 1969, Gerloff and Krombholz 1966, EPA 2002c).
2686
2687 1. The most appropriate plant parts to sample and analyze should be determined. It is
2688 generally recognized that the plant or plant parts should be of the same physiological age.
2689
2690 2. Samples from the same species should be collected and analyzed. Different species
2691 assimilate and concentrate nutrients to different levels.
2692
2693 3. Tissue nutrient concentrations vary with (leaf) position, plant part and age. It is important
2694 to sample and analyze leaves from the same position and age (e.g., third leaf from the
2695 terminal bud on the plant) to ensure comparability of results from sampling of different
2696 wetlands.
2697
2698 4. Tissue P may be a more reliable indicator of nutrient condition than N. This is because N
2699 is used to increase production of aboveground biomass whereas excess P is stored via
2700 luxury uptake.
2701
2702 Another promising macrophyte-based tool is the measurement of nutrient resorption of N and P
2703 prior to leaf senescence and dieback. Nutrient resorption is an important strategy used by
2704 macrophytes to conserve nutrients (Hopkinson and Schubauer 1984; Shaver and Melillo 1984).
2705 In nutrient-poor environments, macrophytes resorb N and P from green leaves prior to
77
-------
December 2006- DRAFT Chapter 5. Candidate Variables for Establishing Nutrient Criteria
2706 senescence, leading to low concentrations of N and P in senesced leaves. In nutrient-rich
2707 environments, resorption becomes less important so that senesced leaves retain much of the N
2708 and P that was present when the leaves were green.
2709
2710 Nitrogen and phosphorus should be measured in green leaves of the same approximate age
2711 collected from the dominant wetland plant species. Samples also should be collected throughout
2712 the wetland to account for spatial variability. If an environmental gradient is known or suspected
2713 to exist within the wetland, then sites along this gradient should be sampled separately. At each
2714 sampling location, approximately five green leaves are collected from each of dominant plant
2715 species. Leaves are collected from the middle portion of the stem, avoiding very young leaves at
2716 the top of the stem and very old leaves at the bottom of the stem. At each location, leaf samples,
2717 by species, are combined for analysis, oven-dried at 70°C and ground.
2718
2719 Nitrogen is measured by dry combustion using a CHN analyzer. Phosphorus is measured
2720 colorimetrically after digestion in strong acid (H2SO4-H2O2) (Allen et al. 1986). Many land-grant
2721 universities, state agricultural testing laboratories, and environmental consulting laboratories
2722 perform these analyses. Contact your local U.S. Department of Agriculture office or land-grant
2723 agricultural extension office for information on laboratories that perform plant tissue nutrient
2724 analyses.
2725
2726 Please see the EPA module 16, Vegetation-based Indicators of Wetland Nutrient Enrichment
2727 (http://www.epa.gov/waterscience/criteria/wetlands/16Indicators.pdf) for a detailed description
2728 of indicators derived from to N and P content of macrophytes.
2729
2730 ABOVEGROUND BIOMASS AND STEM HEIGHT
2731
2732 Wetland macrophytes also respond to nutrient enrichment by increased net primary production
2733 (NPP) and growth if other factors such as light are not limiting growth (Chiang et al. 2000). Net
2734 primary production is the amount of carbon fixed during photosynthesis that is incorporated into
2735 new leaves, stems and roots. Most techniques to measure NPP focus on aboveground biomass
2736 and discount root production because it is difficult to measure even though root production may
2737 account for 50% of NPP. The simplest way to measure aboveground biomass is by harvesting all
2738 of the standing material (biomass) at the end of the growing season (Broome et al., 1986). The
2739 harvest method is useful for measuring NPP of herbaceous emergent vegetation, especially in
2740 temperate climates where there is a distinct growing season. If root production desired, it can be
2741 determined by sequentially harvesting roots at monthly intervals during the year (Valiela et. al,
2742 1976).
2743
2744 Enhanced NPP often is reflected by increased height and, sometimes, stem density of herbaceous
2745 emergent vegetation (Broome et al 1983). Increased stem density, however, may reflect other
2746 factors like vigorous clonal growth so it is not recommended as an indicator of nutrient
2747 enrichment.
78
-------
December 2006- DRAFT Chapter 5. Candidate Variables for Establishing Nutrient Criteria
2748
2749 Aboveground biomass of herbaceous vegetation may be determined by end-of-season harvest of
2750 aboveground plant material in small 0.25 m2 quadrats stratified by macrophyte species or
2751 inundation zone (Broome et al. 1986). Stem height of individuals of dominant species is
2752 measured in each plot. Height of the 5 to 10 tallest stems in each plot has been shown to be a
2753 reliable indicator of NPP (Broome et al. 1986) that saves time as compared to height
2754 measurements of all stems in the plot. Aboveground biomass is clipped at the end of the growing
2755 season, in late summer or fall. Clipped material is separated into live (biomass) versus dead
2756 material then dried at 70°C to a constant weight. For stem height and biomass sampling, 5 to 10
2757 plots per vegetation zone are collected. In forested sites, biomass production is defined as the
2758 sum of the leaf and fruit fall and aboveground wood production (Newbould, 1967). Please see
2759 the EPA module Vegetation-based Indicators of Wetland Nutrient Enrichment
2760 (http://www.epa.gov/waterscience/criteria/wetlands/16Indicators.pdf) for a detailed description
2761 for sampling aboveground biomass in wetlands.
2762
2763 ALGAL NITROGEN & PHOSPHORUS
2764
2765 In some cases, measurements of algal N and P can provide a useful complement to vegetation
2766 and soil nutrient analyses that integrate nutrient history over a period of months in the case of
2767 vegetation (Craft et al. 1995) to years in the case of soils (Craft and Richardson 1998, Chiang et
2768 al. 2000). Nutrient concentrations in algae can integrate variation in water column N and P
2769 bioavailability over a time scale of weeks, potentially providing an indication of the recent
2770 nutrient status of a wetland (Fong et al., 1990; Stevenson et al. 2001; ). Caution is warranted for
2771 this method because it is not useful in all wetlands, for example in wetlands where surface
2772 inundation occurs intermittently or for short periods of time, where the water surface is severely
2773 shaded as in some forested wetlands, or under other circumstances where unrelated
2774 environmental factors exert primary control over algal growth.
2775
2776 Algae should be sampled by collecting grab samples from different locations in the wetland to
2777 account for spatial variability in the wetland. If an environmental gradient is known or suspected
2778 (i.e., decreasing canopy or impacted land uses), or exists within the wetland as a result of
2779 specific source discharges, then sites along this gradient should be sampled separately.
2780 Comparisons among wetlands or locations within a wetland should be done on a habitat-specific
2781 basis (e.g.,phytoplankton versus periphyton). Samples are processed in the same manner as
2782 wetland plants to determine N and P content. Nitrogen is determined using a CHN analyzer
2783 whereas P is measured colorimetrically after acid digestion.
2784
2785 Please see the EPA module Using Algae to Assess Environmental Conditions in Wetlands
2786 (http://www.epa.gov/waterscience/criteria/wetlands/llAlgae.pdf) for a detailed description of
2787 indicators derived from to N and P content of algae.
2788
2789 MACROPHYTE COMMUNITY STRUCTURE AND COMPOSITION
79
-------
December 2006- DRAFT Chapter 5. Candidate Variables for Establishing Nutrient Criteria
2790
2791 The composition of the plant community and the changes that result from human activities can
2792 be used as sensitive indicators of the biological integrity of wetland ecosystems. In particular,
2793 aggressive, fast-growing species such as cattail (Typha spp.), giant reed (Phragmites communis)
2794 reed canarygrass {Phalaris arundincea) and other clonal species invade and may eventually
2795 come to dominate the macrophyte community. Data collection methods and analyses for using
2796 macrophyte community structure and composition as an indicator of nutrient enrichment and
2797 ecosystem integrity for wetlands are described in Vegetation-based Indicators of Wetland
2798 Nutrient Enrichment (http://www.epa.gov/waterscience/criteria/wetlands/16Indicators.pdf) and
2799 Using Vegetation to Assess Environmental Conditions in Wetlands
2800 (http://www.epa.gov/waterscience/criteria/wetlands/10Vegetation.pdf), respectively.
2801
2802 ALGAL COMMUNITY STRUCTURE AND COMPOSITION
2803
2804 Algae can be used as a valuable indicator of biological and ecological condition of wetlands.
2805 Structural and functional attributes of algae can be measured including diversity, biomass,
2806 chemical composition, productivity, and other metabolic functions. Species composition of
2807 algae, particularly of the diatoms, is commonly used as an indicator of biological integrity and
2808 physical and chemical conditions of wetlands. Discussions of sampling, data analyses, and
2809 interpretation are included in Using Algae to Assess Environmental Conditions in Wetlands
2810 (http://www.epa.gov/waterscience/criteria/wetlands/HAlgae.pdf).
2811
2812 INVERTEBRATE COMMUNITY STRUCTURE AND COMPOSITION
2813
2814 Aquatic invertebrates can be used to assess the biological and ecological condition of wetlands.
2815 The approach for developing an Index of Biological Integrity (IBI) for wetlands based on aquatic
2816 invertebrates is described in Developing an Invertebrate Index of Biological Integrity for
2817 Wetlands (http://www.epa.gov/waterscience/criteria/wetlands/9Invertebrate.pdf).
2818
2819 SUMMARY
2820
2821 Candidate variables to use in determining nutrient condition of wetlands and to help identify
2822 appropriate nutrient criteria for wetlands consist of supporting variables, causal variables, and
2823 response variables. Supporting variables provide information useful in normalizing causal and
2824 response variables and categorizing wetlands. Causal variables are intended to characterize
2825 nutrient availability (or assimilation) in wetlands and could include nutrient loading rates and
2826 soil nutrient concentrations. Response variables are intended to characterize biotic response and
2827 could include community structure and composition of macrophytes and algae.
2828
2829 The complex temporal and spatial structure of wetlands will influence the selection of variables
2830 to measure and methods for measuring them. The information contained in this chapter is a brief
2831 summary of suggested analyses that can be used to determine wetland condition with respect to
80
-------
December 2006- DRAFT Chapter 5. Candidate Variables for Establishing Nutrient Criteria
2832 nutrient status. The authors recognize that the candidate variables and analytical methods
2833 described here will generally be the most useful to identifying wetland nutrient condition, other
2834 methods and analyses may be more appropriate in certain systems.
2835
81
-------
December 2006 - DRAFT Chapter 6. Database Development and New Data Collection
2836 Chapter 6 Database Development and New Data
283? Collection
2838
2839 6.1 INTRODUCTION
2840
2841 A database of relevant water quality information can be an invaluable tool to States and Tribes as
2842 they develop nutrient criteria. In some cases existing data are available and can provide
2843 additional information that is specific to the region where criteria are to be set. However, little or
2844 no data are available for most regions or parameters, and creating a database of newly gathered
2845 data is strongly recommended. In the case of existing data, the data should be located, and their
2846 suitability (type and quality and sufficient associated metadata) ascertained. It is also important
2847 to determine how the data were collected to ensure that future monitoring efforts are compatible
2848 with earlier approaches.
2849
2850 Databases operate much like spreadsheet applications, but have greater capabilities. Databases
2851 store and manage large quantities of data and allow viewing and exporting of data sorted in a
2852 variety of ways, while spreadsheets analyze and graphically display small quantities of data.
2853 Databases can be used to organize existing information, store newly gathered monitoring data,
2854 and manipulate data for water quality criteria development. Databases can sort data for export
2855 into statistical analyses programs, spreadsheets, and graphics programs. This chapter will discuss
2856 the role of databases in nutrient criteria development, and provide a brief review of existing
2857 sources of nutrient-related water quality information for wetlands.
2858
2859
2860 6.2 DATABASES AND DATABASE MANAGEMENT
2861 A database is a collection of information related to a particular subject or purpose. Databases are
2862 arranged so that individual values are kept separate, yet can be linked to other values based on
2863 some common denominator (such as association of time or location). Geographic Information
2864 Systems (GIS) are geo-referenced relational databases that have a geographical component (i.e.,
2865 spatial platform) in the user interface. Spatial platforms associated with a database allow
2866 geographical display of sets of sorted data. GIS platforms such as Arc View™, Arclnfo™, and
2867 Maplnfo™ are frequently used to integrate spatial data with monitoring data for watershed
2868 analysis. Data stored in simple tables, relational database or geo-reference databases can also be
2869 located, retrieved and manipulated using queries. A query allows the user to find and retrieve
2870 only the data that meets user-specified conditions. Queries can also be used to update or delete
2871 multiple records simultaneously and to perform built-in or custom calculations of data. Data in
2872 tables can be analyzed and printed in specific layouts using reports. Data can be analyzed or
2873 presented in a specific way in print by creating a report. The most effective use of these tools
2874 requires a certain amount of training, expertise, and software support, especially when using geo-
2875 referenced data.
82
-------
December 2006 - DRAFT Chapter 6. Database Development and New Data Collection
2876
2877 To facilitate data storage, manipulation and calculations, it is highly recommended that historical
2878 and present-day data be transferred to a relational database (i.e., Access™). Relational
2879 databases store data in tables as sets of rows and columns, and are powerful tools for data
2880 manipulation and initial data reduction. They allow selection of data by specific, multiple
2881 criteria, and definition and redefinition of linkages among data components. Data queries can
2882 also be exported to GIS provided data is related to some geo-referenced coordinate system.
2883
2884 POTENTIAL DATA SOURCES
2885
2886 EPA Water Quality Data
2887
2888 STORET
2889 EPA has many programs of national scope that focus on collection and analysis of water quality
2890 data. The following presents information on several of the databases and national programs that
2891 may be useful to water quality managers as they compile data for criteria development.
2892 STORET STOrage and RETrieval system (STORET) is EPA's national database for water
2893 quality and biological data.
2894
2895 Environmental Monitoring and Assessment Program (EMAP)
2896 The Environmental Monitoring and Assessment Program is an EPA research program designed
2897 to develop the tools necessary to monitor and assess the status and trends of national ecological
2898 resources (see EMAP Research Strategy on the EMAP website: www.epa.gov/emap). EMAP's
2899 goal is to develop the scientific understanding for translating environmental monitoring data
2900 from multiple spatial and temporal scales into assessments of ecological condition and forecasts
2901 of future risks to the sustainability of the Nation's natural resources. Data from the EMAP
2902 program can be downloaded directly from the EMAP website (www.epa.gov/emap/). The EMAP
2903 Data Directory contains information on available data sets including data and metadata
2904 (language that describes the nature and content of data). Current status of the data directory as
2905 well as composite data and metadata files are available on this website.
2906
2907 USGS (U.S. Geological Survey) Water Data
2908
2909 The USGS has national and distributed databases on water quantity and quality for waterbodies
2910 across the nation. Much of the data for rivers and streams are available through the National
2911 Water Information System (NWIS). These data are organized by state, Hydrologic Unit Codes
2912 (HUCs), latitude and longitude, and other descriptive attributes. Most water quality chemical
2913 analyses are associated with an instantaneous streamflow at the time of sampling and can be
2914 linked to continuous streamflow to compute constituent loads or yields. The most convenient
2915 method of accessing the local data bases is through the USGS State representative. Every State
2916 office can be reached through the USGS home page at: http://www.usgs.gov.
2917
83
-------
December 2006 - DRAFT Chapter 6. Database Development and New Data Collection
2918 HBN and NASQAN
2919 USGS data from several national water quality programs covering large regions offer highly
2920 controlled and consistently collected data that may be particularly useful for nutrient criteria
2921 analysis. Two programs, the Hydrologic Benchmark Network (HBN) and the National Stream
2922 Quality Accounting Network (NASQAN) include routine monitoring of rivers and streams
2923 during the past 30 years. The HBN consisted of 63 relatively small, minimally disturbed
2924 watersheds. HBN data were collected to investigate naturally-induced changes in streamflow and
2925 water quality and the effects of airborne substances on water quality. The NASQAN program
2926 consists of 618 larger, more culturally influenced watersheds. NASQAN data provides
2927 information for tracking water-quality conditions in major U.S. rivers and streams. The
2928 watersheds in both networks include a diverse set of climatic, physiographic, and cultural
2929 characteristics. Data from the networks have been used to describe geographic variations in
2930 water-quality concentrations, quantify water-quality trends, estimate rates of chemical flux from
2931 watersheds, and investigate relations of water quality to the natural environment and
2932 anthropogenic contaminant sources.
2933
2934 WEBB
2935 The Water, Energy, and Biogeochemical Budgets (WEBB) program was developed by USGS to
2936 study water, energy, and biogeochemical processes in a variety of climatic/regional scenarios.
2937 Five ecologically diverse watersheds, each with an established data history, were chosen. This
2938 program may prove to be a rich data source for ecoregions in which the five watersheds are
2939 located. Many publications on the WEBB project are available. See the USGS website for more
2940 details (http://water.usgs.gov/nrp/webb/about.html).
2941
2942 US Department of Agriculture (USDA)
2943 Agricultural Research Service (ARS)
2944
2945 The USDA ARS houses the Natural Resources and Sustainable Agricultural Systems Scientific
2946 Directory (http://hydrolab.arsusda.gov/arssci.html), which has seven national programs to
2947 examine the effect of agriculture on the environment. The program on Water Quality and
2948 Management addresses the role of agriculture in nonpoint source pollution through research on
2949 Agricultural Watershed Management and Landscape Features, Irrigation and Drainage
2950 Management Systems, and Water Quality Protection and Management Systems. Research is
2951 conducted across the country and several models and databases have been developed.
2952 Information on research and program contacts is listed on the website
2953 (http://www.nps.ars.usda.gov/programs/nrsas.htm).
2954
2955 Forest Service
2956
2957 The Forest Service has designated research sites across the country, many of which are Long
2958 Term Ecological Research (LTER) sites. Many of the data from these experiments are available
2959 in the USFS databases located on the website (http://www.fs.fed.us/research/). Most of the data
84
-------
December 2006 - DRAFT Chapter 6. Database Development and New Data Collection
2960 are forest-related, but may be of use for determining land uses and questions on silviculture
2961 runoff.
2962
2963 National Science Foundation (NSF)
2964
2965 The National Science Foundation (NSF) funds projects for the Long Term Ecological Research
2966 (LTER) Network. The Network is a collaboration of over 1,100 researchers investigating a wide
2967 range of ecological topics at 24 different sites nationwide. The LTER research programs are not
2968 only an extremely rich data source, but also a source of data available to anyone through the
2969 Network Information System (NTS), the NSF data source for LTER sites. Data sets from sites are
2970 highly comparable due to standardization of methods and equipment.
2971
2972 U.S. Army Corps of Engineers (COE)
2973
2974 The U.S. Army Corps of Engineers (COE) is responsible for many federal wetland jurisdiction
2975 issues. Although a specific network of water quality monitoring data does not exist, specific
2976 studies on wetlands by the COE may provide suitable data. The COE focuses more on water
2977 quantity issues than on water quality issues. As a result, much of the wetland system data
2978 collected by the COE does not include nutrient data. Nonetheless, the COE does have a large
2979 water sampling network and supports USGS and EPA monitoring efforts in many programs. A
2980 list of the water quality programs that the COE actively participates in can be found at
2981 http://www.usace.army.mil/public.html.
2982
2983 U.S. Department of the Interior, Bureau of Reclamation (BuRec)
2984
2985 The Bureau of Reclamation of the US Department of the Interior manages many irrigation and
2986 water supply reservoirs in the West, some of which may have wetland applicable data available.
2987 These data focus on water supply information and limited water quality data. However, real time
2988 flow data are collected for rivers supplying water to BuRec, which may be useful if a flow
2989 component of criteria development is chosen. These data can be gathered on a site-specific basis
2990 from the BuRec website: http://www.usbr.gov.
2991
2992 State/Tribal Monitoring Programs
2993
2994 Some states may have wetland water quality data as part of a research study, use attainability
2995 analysis (UAA), or to assess mitigation or nutrient related impacts. Most of this data is collected
2996 by State natural resources or environmental protection agencies, or by regional water
2997 management authorities. Data collected by State/Tribal water quality monitoring programs can
2998 be used for nutrient criteria development and may provide pertinent data sources although they
2999 may be regionally limited. These data should be available from the agencies responsible for
3000 monitoring.
3001
85
-------
December 2006 - DRAFT Chapter 6. Database Development and New Data Collection
3002 Volunteer Monitoring Programs
3003
3004 State and local agencies may use volunteer data to screen for water quality problems, establish
3005 trends in waters that would otherwise be unmonitored, and make planning decisions. Volunteers
3006 benefit from learning more about their local water resources and identifying what conditions or
3007 activities might contribute to pollution problems. As a result, volunteers frequently work with
3008 clubs, environmental groups, and State/Tribal or local governments to address problem areas.
3009 The EPA supports volunteer monitoring and local involvement in protecting our water resources.
3010
3011 Academic and Literature Sources
3012
3013 Most of the data available on water and soil quality in wetlands is the result of research studies
3014 conducted by academic institutions. Much of the research conducted by the academic
3015 community, however, was not conducted for the purpose of spatial or long-term biogeochemical
3016 characterization of the nation's wetlands; instead water quality information was often collected
3017 to characterize the environmental conditions under which a particular study or experiment was
3018 conducted. Infrequently spatial studies of limited spatial extent or duration were conducted. Data
3019 collected from these sources therefore, may not be sufficiently representative of the population
3020 of wetlands within an ecoregion. However, this limited data may be the only information
3021 available and therefore could be useful for identifying reference conditions or where to begin a
3022 more comprehensive survey to support development of nutrient criteria. Academic research data
3023 is available from researchers and the scientific literature.
3024
3025
3026 6.3 QUALITY OF HISTORICAL AND COLLECTED DATA
3027
3028 The value of older historical data is a recurrent problem because data quality is often unknown.
3029 Knowledge of data quality is also problematic for long-term data repositories such as STORET
3030 and long-term State databases, where objectives, methods, and investigators may have changed
3031 many times over the years. The most reliable data tend to be those collected by a single agency
3032 using the same protocol. Supporting documentation should be examined to determine the
3033 consistency of sampling and analytical protocols. The suitability of data in large, heterogeneous
3034 data repositories for establishing nutrient criteria are described below. These same factors need
3035 to be taken into account when developing a new database such that future investigators will have
3036 sufficient information necessary to evaluate the quality of the database.
3037
3038 LOCATION
3039
3040 Geo-referenced data is extremely valuable in that it allows for aggregating and summarizing data
3041 according to any GIS coverage desired, whether the data was historically related to a particular
3042 coverage theme or not. However, many studies conducted prior to the availability and accuracy
3043 of hand held Global Positioning System (GPS) units relied on narrative and less definitive
86
-------
December 2006 - DRAFT Chapter 6. Database Development and New Data Collection
3044 descriptions of location such as proximity to transportation corridor, county or nearest municipal
3045 center. This can make comparison of data, depending upon desired spatial resolution, difficult.
3046 Knowledge of the rationale and methods of site selection from the original investigators may
3047 supply valuable information to determine whether inclusion of the site or study in the database is
3048 appropriate based on potential bias relative to overall wetland data sources. STORET and USGS
3049 data associated with the National Hydrography Dataset (NHD) are geo-referenced with latitude,
3050 longitude, and Reach File 3 (RF3) codes (http://nhd.usgs.gov/). In addition, STORET often
3051 contains a site description to supplement location information. Metadata of this type, when
3052 known, is frequently stored within large long-term databases.
3053
3054 VARIABLES AND ANALYTICAL METHODS
3055
3056 Each separate analytical method yields a unique variable. For example, five ways of measuring
3057 TP results in five unique variables. Data generated using different analytical methods should not
3058 be combined in data analyses because methods differ in accuracy, precision, and detection limits.
3059 Data generated from one method may be too limited, making it important to select the most
3060 frequently used analytical methods in the database. Data that were generated using the same
3061 analytical methods may not always be obvious because of synonymous names or analytical
3062 methods. Consistency in taxonomic conventions and indicator measurements is likewise
3063 important for biological variables and multimetric indices comparisons. Review of recorded data
3064 and analytical methods by knowledgeable personnel is important to ensure that there are no
3065 problems with datasets developed from a particular database.
3066
3067 LABORATORY QUALITY CONTROL (QC)
3068
3069 Data generated by agencies or laboratories with known quality control/quality assurance
3070 protocols are most reliable. Laboratory QC data (blanks, spikes, replicates, known standards) are
3071 infrequently reported in larger data repositories. Records of general laboratory quality control
3072 protocols and specific quality control procedures associated with specific datasets are valuable in
3073 evaluating data quality. However, premature elimination of lower quality data can be
3074 counterproductive, because the increase in variance caused by analytical laboratory error may be
3075 negligible compared to natural variability or sampling error, especially for nutrients and related
3076 water quality parameters. However, data of uncertain and undocumented quality should not be
3077 accepted.
3078
3079 Water column nutrient data can be reported in different units, e.g.,ppm, mg/L, mmoles.
3080 Reporting of nutrient data from other strata such as soils, litter and vegetation can further expand
3081 the list of reporting units (e.g.,mg/kg, g/kg, %, mg/cm3). In many instances conversion of units is
3082 possible, however, in other instances unit conversion is not possible or is lacking support
3083 information for conversion. Consistency in reporting units and the need to provide conversion
3084 tables cannot be overemphasized.
3085
87
-------
December 2006 - DRAFT Chapter 6. Database Development and New Data Collection
3086 DATA COLLECTING AGENCIES
3087
3088 Selecting data from particular agencies with known, consistent sampling and analytical methods
3089 and known quality will reduce variability due to unknown quality problems. Requesting data
3090 review for quality assurance from the collecting agency will reduce uncertainty about data
3091 quality.
3092
3093 TIME PERIOD
3094
3095 Long-term records are critically important for establishing trends. Determining if trends exist in
3096 the time series database is also important for characterizing reference conditions for nutrient
3097 criteria. Length of time series data needed for analyzing nutrient data trends is discussed in
3098 Chapter 7.
3099
3100 INDEX PERIOD
3101
3102 An index period-the time period most appropriate for sampling-for estimating average
3103 concentrations can be established if nutrient and water quality variables were measured through
3104 seasonal cycles. The index period may be the entire year or the summer growing season. The
3105 best index period is determined by considering wetland characteristics for the region, the quality
3106 and quantity of data available, and estimates of temporal variability (if available). Consideration
3107 of the data available relative to longer-term oscillations in environmental conditions, e.g., dry
3108 years, wet years, should also be taken into account such that the data is representative and
3109 appropriate. Additional information and considerations for establishing an index period are
3110 discussed in Chapter 7.
3111
3112 REPRESENTATIVENESS
3113
3114 Data may have been collected for specific purposes. Data collected for toxicity analyses, effluent
3115 limit determinations, or other pollution problems may not be useful for developing nutrient
3116 criteria. Further, data collected for specific purposes may not be representative of the region or
3117 wetland classes of interest. The investigator should determine if all wetlands or a subset of the
3118 wetlands in the database are representative of the population of wetlands to be characterized. If a
3119 sufficient sample of representative wetlands cannot be found, then a new survey is strongly
3120 recommended.
3121
3122
3123 6.4 COLLECTING NEW DATA
3124
3125 New data should be collected when no data presently exist or the data available are not suitable,
3126 and should be gathered following the sampling design protocols discussed in Chapter 4. New
3127 data collection activities for developing nutrient criteria should focus on filling in gaps in the
-------
December 2006 - DRAFT Chapter 6. Database Development and New Data Collection
3128 database and collecting spatially representative regional monitoring data. In many cases this may
3129 mean starting from scratch because no data presently exists or that the data available are not
3130 suitable. Data gathered under new monitoring programs should be imported into databases or
3131 spreadsheets and, if comparable, merged with existing data for criteria development. It is best to
3132 archive the data with as much data-unique information (meta-data) as possible. It is always
3133 possible to aggregate at a later time, but impossible to separate lumped data without having the
3134 parameter needed to partition the dataset. Redundancy may also be a problem, but can more
3135 easily be avoided when common variables or parameters are kept in each database (i.e., dates
3136 may be very important). The limitations and qualifications of each data set should be known, and
3137 data 'tagged' if possible, before combining them. The following five factors should be
3138 considered when collecting new data and before combining new data with existing data sets:
3139 representativeness, completeness, comparability, accuracy and precision.
3140
3141 REPRESENTATIVENESS
3142
3143 Sampling program design (when, where, and how you sample) should produce samples that are
3144 representative or typical of the regional area being described and the classes of wetlands present.
3145 Sampling designs for developing nutrient criteria are addressed in Chapter 4. Databases
3146 populated by data from the literature or historical studies will not likely provide sufficient spatial
3147 or class representation of a region. Data interpretation should recognize these gaps and should be
3148 limited until gaps are filled using additional survey information.
3149
3150 COMPLETENESS
3151
3152 A QA/QC plan should describe how to complete the data set in order to answer questions posed
3153 (with a statistical test of given power and confidence) and the precautions being taken to ensure
3154 that completeness. Data collection procedures should document the extent to which these
3155 conditions have been met. Incomplete data sets may not invalidate the collected data, but may
3156 reduce the rigor of statistical analyses. Precautions to ensure completeness may include
3157 collecting extra samples, having back-up equipment in the field, copying field notebooks after
3158 each trip, and/or maintaining duplicate sets of data in two locations.
3159
3160 COMPARABILITY
3161
3162 In order to compare data collected under different sampling programs or by different agencies,
3163 sampling protocols and analytical methods should demonstrate comparable data. The most
3164 efficient way to produce comparable data is to use sampling designs and analytical methods that
3165 are widely used and accepted, and examined for compatibility with other monitoring programs
3166 prior to initiation of a survey. Comparability should be assessed for field sample collection,
3167 sample preservation, sample preparation and analysis, and among laboratories used for sample
3168 analyses.
3169
89
-------
December 2006 - DRAFT Chapter 6. Database Development and New Data Collection
3170 ACCURACY
3171
3172 To assess the accuracy of field instruments and analytical equipment, a standard (a sample with a
3173 known value) should be analyzed and the measurement error or bias determined. Internal
3174 standards should periodically be checked with external standards provided by acknowledged
3175 sources. At Federal, State, Tribal, and local government levels, the National Institute of
3176 Standards and Technology (NIST) provides advisory and research services to all agencies by
3177 developing, producing, and distributing standard reference materials for vegetation, soils, and
3178 sediments. Standards and methods of calibration are typically included with turbidity meters, pH
3179 meters DO meters, and DO testing kits. The USEPA, USGS, and some private companies
3180 provide reference standards or QC samples for nutrients.
3181
3182 VARIABILITY
3183
3184 The variability in field measurements and analytical methods should be demonstrated and
3185 documented to identify the source and magnitude of variability when possible. EPA QA/QC
3186 guidance provides an explanation and protocols for measuring sampling variability (USEPA
3187 1998c).
3188
3189 DATA REDUCTION
3190
3191 For data reduction, it is important to have a clear idea of the analysis that will be performed and
3192 a clear definition of the sample unit for analysis. For example, a sample unit might be defined as
3193 "a wetland during July- August". For each variable measured, a mean value would then be
3194 estimated for each wetland during the July-August index period on record. Analyses are then
3195 conducted on the observations (estimated means) for each sample unit, not with the raw data.
3196 Steps recommended for reducing the data include:
3197
3198 1. Selecting the long-term time period for analysis;
3199
3200 2. Selecting an index period;
3201
3202 3. Selecting relevant variables of interest;
3203
3204 4. Identifying the quality of analytical methods;
3205
3206 5. Identifying the quality of the data recorded; and
3207
3208 6. Estimating values for analysis (mean, median, minimum, maximum) based on the
3209 reduction selected.
3210
3211
90
-------
December 2006 - DRAFT Chapter 6. Database Development and New Data Collection
3212 6.5 QUALITY ASSURANCE / QUALITY CONTROL (QA/QC)
3213
3214 The validity and usefulness of data depend on the care with which they were collected, analyzed
3215 and documented. EPA provides guidance on data quality assurance (QA) and quality control
3216 (QC) (USEPA 1998c) to assure the quality of data. Factors that should be addressed in a QA/QC
3217 plan are elaborated below. The QA/QC plan should state specific goals for each factor and
3218 should describe the methods and protocols used to achieve the goals.
3219
3220 1. Who will use the data?
3221 2. What the project's goals/objectives/questions or issues are?
3222 3. What decision(s) will be made from the information obtained?
3223 4. How, when, and where project information will be acquired or generated?
3224 5. What possible problems may arise and what actions can be taken to mitigate their impact
3225 on the project?
3226 6. What type, quantity, and quality of data are specified?
3227 7. How "good" those data have to be to support the decision to be made?
3228 8. How the data will be analyzed, assessed, and reported?
3229
3230
3231
3232
91
-------
December 2006 - DRAFT Chapter 7. Data Analysis: Experiments, Observational Data, and Modeling
3233 Chapter 7 Data Analysis
3234
3235
3236 7.1 INTRODUCTION
3237
3238 Data analysis is critical to nutrient criteria development. Proper analysis and interpretation of
3239 data determine the scientific defensibility and effectiveness of the criteria. Therefore, it is
3240 important to evaluate short and long-term goals for wetlands of a given class within the region of
3241 concern. These goals should be addressed when analyzing and interpreting nutrient and response
3242 data. Specific objectives to be accomplished through use of nutrient criteria should be identified
3243 and revisited regularly to ensure that goals are being met. The purpose of this chapter is to
3244 explore methods for analyzing data that can be used to develop nutrient criteria consistent with
3245 these goals. Included are techniques to evaluate metrics, to examine or compare distributions of
3246 nutrient exposure or response variables, and to examine nutrient exposure-response
3247 relationships.
3248
3249 Statistical analyses are used to interpret monitoring data for criteria development. Statistical
3250 methods are data-driven, and range from very simple descriptive statistics to more complex
3251 statistical analyses. Generally, the type of statistical analysis used for criteria development is
3252 determined by the source, quality, and quantity of data available.
3253
3254
3255 7.2 FACTORS AFFECTING ANALYSIS APPROACH
3256 Wetland systems should be appropriately classified a priori for nutrient criteria development to
3257 minimize natural background variation (see Chapter 3). This section discusses some of the
3258 factors that should be considered when classifying wetland systems, and in determining the
3259 choice of predictor (causal) and response variables to include in the analysis.
3260
3261 Wetland hydrogeomorphic type http://el.erdc.usace.army.mil/wrap/wrap.html may determine the
3262 sensitivity of wetlands to nutrient inputs, as well as the interaction of nutrients with other driving
3263 factors in producing an ecological response. Hydrogeomorphic types differ in landscape
3264 position, predominant water source, and hydrologic exchanges with adjacent water bodies
3265 (Brinson 1993). These factors in turn influence water residence time, hydrologic regime, and
3266 disturbance regime. In general, isolated depressional wetlands will have greater residence times
3267 than fringe wetlands, which in turn will have greater residence times than riverine wetlands.
3268 Systems with long residence times are likely to behave more like lakes than flow-through
3269 systems, and may show a greater response to cumulative loadings. Thus, nutrient loading rates or
3270 indicators thereof are likely to be a more sensitive predictor of ecological effects for depressional
3271 wetlands, while nutrient water column or sediment concentrations are likely to be a more
3272 sensitive predictor of responses for riverine wetlands. Water column concentrations will
3273 influence the response of algal communities, while macrophytes derive nutrients from both the
92
-------
December 2006 - DRAFT Chapter 7. Data Analysis: Experiments, Observational Data, and Modeling
3274 water column and sediments. Fringe wetlands are likely to be influenced both by concentration
3275 of nutrients in the adjacent lake or estuary as well as the accumulation of nutrients within these
3276 systems from groundwater inflow and, in some cases, riverine inputs. The relative influence of
3277 these two sources will depend on the exchange rate with the adjacent lake, e.g., through seiche
3278 activity (Keough et al., 1999; Trebitz et al., 2002). In practice, it is difficult to measure loadings
3279 from multiple sources including groundwater and exchange with adjacent water bodies. If
3280 sediment concentrations are shown to be a good indicator of recent loading rates, then sediment
3281 concentrations might be the best predictor to use across systems.
3282
3283 It may be important to control for ancillary factors when teasing out the relationship between
3284 nutrients and vegetation community response, particularly if those factors interact with nutrients
3285 in eliciting responses. For example, riverine and fringe wetlands differ from basin wetlands in
3286 the frequency and intensity of disturbance from flooding events or ice. Day et. al. (1988)
3287 describe a fertility-disturbance gradient model for riverine wetlands describing how the relative
3288 dominance of plant guilds with different growth forms and life history strategies depends on the
3289 interactive effects of productivity, fertility, disturbance, and water level. In depressional
3290 wetlands, the model could be simplified to include only the interaction of fertility with the
3291 hydrologic regime. Disturbance regimes and water level could be incorporated into analysis of
3292 cause-effect relationships either as categorical factors or as covariates.
3293
3294 The selection of assessment and measurement of response attributes for determining ecological
3295 response to nutrient loadings should depend, in part, on designated uses assigned to wetlands as
3296 part of standards development. Designated uses such as recreation (aesthetics and contact) or
3297 drinking water are not typically assigned to wetlands; thus defining nuisance algal blooms in
3298 terms of taste or odor problems or aesthetic considerations may not be appropriate for wetlands.
3299 Guidance for the definition of aquatic life use is currently being refined to describe six stages of
3300 impact along a human disturbance gradient, from pristine reference condition to heavily
3301 degraded sites (Figure 7, Stevenson and Hauer 2002, Davies and Jackson 2006). The relative
3302 abundance of sensitive native taxa is expected to shift with relatively minor impacts, while
3303 organism condition or functional attributes are relatively robust to altered loadings. However, if
3304 maintenance of ecological integrity of sensitive downstream systems is of concern, then it may
3305 be important to measure some functional attributes related to nutrient retention. Stevenson and
3306 Hauer (2002) have suggested a series of "resource condition tiers" analogous to those defined for
3307 biological condition, but related to ecosystem functions. Tier 1 requirements are proposed as:
3308 "Native structure and function of the hydrologic and geomorphic regimes and processes are in
3309 the natural
3310
93
-------
December 2006 - DRAFT
Chapter 7. Data Analysis: Experiments, Observational Data, and Modeling
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
c
o
C.
O
O
_o
o
m
BCG Model: Snap Shot
Natural structure & function of biotic community maintained
Minimal changes in structure & function
Evident changes in structure and
minimal changes in function
Moderate changes in structure &
minimal changes in function
Major changes in structure &
moderate changes in function
Severe changes in structure & function
Increasing Levels of Stressors
Figure 7.1. Biological condition gradient model describing biotic community condition as levels
of stressors increase.
range of variation in time and space." Thus maintenance of structure and function of upstream
processes should be protective of downstream biological conditions.
7.3 DISTRIBUTION-BASED APPROACHES
Frequency distributions can aid in the setting of criteria by describing central tendency and
variability among wetlands. Approaches to numeric nutrient criteria development based on
frequency distributions do not require specific knowledge of individual wetland condition prior
to setting criteria using frequency distributions. Criteria are based on and, in a sense, developed
relative to the conditions of the population of wetlands of a given class in the Region, State, or
Tribal lands.
The simplest statistic describing the shape of distributions refers to quartiles, or the 25th and the
75th percentile. These can be defined as the observation which has either 25 % of the
observations on one side and 75 % on the other side in the case of the first quartile (25th
percentile) or vice versa in the case of the third quartile (75th percentile). In the same manner, the
median is the second quartile or the 50th percentile. Graphically, this is depicted in the boxplots
94
-------
December 2006 - DRAFT Chapter 7. Data Analysis: Experiments, Observational Data, and Modeling
3334 as the box length, the lower extreme represents the first quartile, the upper extreme represents
3335 the third quartile, the area inside the box encompassing 50 % of the data.
3336
3337 Distributions of nutrient exposure metrics or response variables can be developed to represent
3338 either an entire population of wetlands, or only a subset of these considered to be minimally
3339 impacted. In either case, a population of wetlands should be defined narrowly enough through
3340 classification so that the range in attributes due to natural variability does not equal or exceed the
3341 range in attributes related to anthropogenic effects. The effects of natural variability can be
3342 minimized by classifying wetlands by type and/or region. Nutrient ecoregions define one
3343 potential regional classification system (USEPA 2000). Alternatively, thresholds in landscape or
3344 watershed attributes defining natural breakpoints in nutrient concentrations can be determined
3345 objectively through procedures such as classification and regression tree (CART) analysis
3346 (Robertson et. al., 2001).
3347
3348
3349 7.4 RESPONSE-BASED APPROACHES
3350
3351 Indicators characterized as "response" or "condition" metrics should be distinguished from
3352 "stressor" or "causal" indicators, such as nutrient concentrations (Paulsen et al., 1991; USEPA
3353 1998a; Stevenson 2004a). While both "response" and "causal" indicators could be used in a
3354 single multimetric index, it is recommended that separate multimetric indices be used for
3355 "response" and "causal" assessment. Distinguishing between "response" and "causal" indices can
3356 be accomplished utilizing a risk assessment approach with separate hazard and exposure
3357 assessments that are linked to response-stressor relationships (USEPA 1996; 1998a; Stevenson
3358 1998; Stevenson et al., 2004a, b). A multimetric index that specifically characterizes "responses"
3359 can be used to clarify goals of management (maintenance or restoration of ecological attributes)
3360 and to measure whether goals have been attained with nutrient management strategies.
3361 Response-based multimetric indices can also be used more directly for natural resource damage
3362 assessments than multimetric indices with response and causal variables.
3363
3364 Factors that should be considered in selecting indicators include conceptual relevance (relevance
3365 to the assessment and to ecological function), feasibility of implementation (data collection
3366 logistics, information management, quality assurance, cost), response variability (measurement
3367 error, seasonal variability, interannual variability, spatial variability, discriminatory ability), and
3368 interpretation and utility (data quality objectives, assessment thresholds, link to management
3369 actions) (Jackson et al., 2000). Of these factors, cost, response variability, and ability to meet
3370 data quality objectives can be assessed through quantitative methods. An analytical
3371 understanding of the factors that affect wetlands the most will also help States and Tribes
3372 develop the most effective monitoring and assessment strategies
3373
3374 Designated uses such as contact recreation and drinking water may not be applicable to
3375 wetlands, hence, it may not be readily apparent what the relative significance of changes in
95
-------
December 2006 - DRAFT Chapter 7. Data Analysis: Experiments, Observational Data, and Modeling
3376 different primary producers is for organisms at higher trophic levels. Wetland food webs have
3377 traditionally been considered to be detritus-based (Odum and de la Cruz 1967; Mann 1972,
3378 1988). However, more recent research on wetland food webs utilizing stable isotope analysis
3379 have identified the importance of phytoplankton, periphyton, or benthic algae as the base of the
3380 food chain for higher trophic levels (Fry 1984, Kitting et al., 1984, Sullivan and Moncreiff 1990,
3381 Hamilton et al., 1992, Newell et al., 1995, Keough et al., 1996); in these cases, it would be
3382 particularly important to monitor shifts in algal producers.
3383
3384 Empirical relationships can be derived directly between water quality parameters such as total P
3385 or transparency and wetland biological responses. Unlike lakes or streams, the level of algal
3386 biomass corresponding to aesthetic problems or ecological degradation in wetlands is not readily
3387 defined, so that defining a TP-chlorophyll a relationship based on water column measurements is
3388 not likely to be useful. However, in some wetlands such as coastal Great Lakes, the loss of
3389 submerged aquatic vegetation biomass and/or diversity with increased eutrophication provides
3390 an ecologically significant endpoint (Lougheed et al., 2001). Reductions in submerged plant
3391 species diversity was associated with increases in turbidity, total P, total N, and chlorophyll a,
3392 suggesting that a trophic state index incorporating multiple parameters might be a better
3393 predictor than a single variable such as total P (Carlson 1977).
3394
3395 Models describing empirical relationships can include linear or nonlinear univariate forms with a
3396 single response metric, multivariate with multiple response metrics, a series of linked
3397 relationships, and simulation models. The simplest forms of linear univariate approaches are
3398 correlation and regression analyses; these approaches have the advantage that they are simple to
3399 perform and transparent to the general public. When assessment thresholds can be determined
3400 based on severity of effect or difference from reference conditions, such that associated exposure
3401 criteria can be derived, linear forms should be adequate. In the case of nonlinear relationships,
3402 data can generally be transformed to linearize the relationship. However, if it is desired to
3403 identify the inflection point in a curvilinear relationship as an indicator of rapid ecological
3404 change, alternative data analysis methods are available, including changepoint analysis
3405 (Richardson and Qian 1999) and piecewise iterative regression techniques (Wilkinson 1999).
3406
3407 Multivariate models are useful for relating nutrient exposure metrics to community-level
3408 responses. Both parametric and nonparametric (nonmetric dimensional scaling orNMDS)
3409 ordination procedures can be used to define axes or gradients of variation in community
3410 composition based on relative density, relative abundance, or simple presence-absence measures
3411 (Gauch 1982, Beals 1984, Heikkila 1987, Growns et al., 1992). Ordination scores then can be
3412 regressed against nutrient exposure metrics, as an indicator of a composite response (McCormick
3413 et al., 1996). Direct gradient analysis techniques such as canonical correspondence analysis can
3414 be used to determine which combination of nutrient exposure variables predict a combination of
3415 nutrient response variables as a first step in deriving multimetric exposure and response variables
3416 (Cooper et al., 1999). Indicator analysis can be used to determine which subset of species best
3417 discriminate between reference sites with low nutrient loadings versus potentially impacted sites
96
-------
December 2006 - DRAFT Chapter 7. Data Analysis: Experiments, Observational Data, and Modeling
3418 with high loadings, or weighted averaging techniques can be used to infer nutrient levels from
3419 species composition (McCormick et al., 1996, Cooper et al., 1999, Jensen et al., 1999). In the
3420 latter case, paleoecological records can be examined to infer historic changes in total P levels
3421 from macrophyte pollen or diatom frustrules, which will be particularly valuable in the absence
3422 of sites representing reference condition (Cooper et al., 1999, Jensen et al.,1999).
3423
3424 Some ecohydrological models have been derived that incorporate the effect of multiple stressors
3425 (hydrology, eutrophication, acidity) on wetland vegetation, thus providing a link between
3426 process-based models and community level response (see Olde Venterink and Wassen 1997 for
3427 review). These models are based on 1) a combination of expert opinion to estimate species
3428 sensitivities, supplemented by multivariate classification of vegetation and environmental data to
3429 determine boundaries of species guilds, or 2) field measurements used to derive logistic models
3430 to quantify dose-response. These approaches could be used to derive wetland nutrient criteria for
3431 the US, provided that models could be calibrated using species and response curves developed
3432 using data for the US. Most multiple-stressor models for wetland vegetation have been calibrated
3433 using data from western Europe (Olde Venterink and Wassen 1997). Latour and colleagues
3434 (Latour and Reiling 1993, Latour et al., 1994) have suggested a mechanism for setting nutrient
3435 standards using the occurrence probability of species along a trophic gradient to extrapolate
3436 maximum tolerable concentrations that protect 95% of species.
3437
3438 A series of linked empirical relationships for wetlands may be most effective for developing
3439 nutrient criteria. Linked empirical relationships may be most useful in cases where integrative
3440 exposure measurements such as sediment nutrient concentrations are more sensitive predictors of
3441 shifts in community composition, or algal P limitation, or other ecological responses
3442 (phosphatase enzyme assays; Qian et al., 2003) than are spatially and temporally heterogeneous
3443 water column nutrient concentrations. In these cases, it may be important to develop one set of
3444 relationships between nutrient loading and exposure indicators for a subset of sites at which
3445 intensive monitoring is done, and another set of relationships between nutrient exposure and
3446 ecological response indicators for a larger sample population (Qian et al., 2003).
3447
3448
3449 7.5 PARTITIONING EFFECTS AMONG MULTIPLE STRESSORS
3450
3451 Changes in nutrient concentrations within or loadings to wetlands often co-occur with other
3452 potential stressors such as changes in hydrologic regime and sediment loading. In a few cases,
3453 researchers have been able to separate the simple effects of nutrient addition through
3454 manipulations of mesocosms (Busnardo et al., 1992, Gabor et al. 1994, Murkin et al., 1994,
3455 McDougal et al., 1997, Hann and Goldsborough 1997), segments of natural systems (Richardson
3456 and Qian 1999, Thormann and Bayley 1997), or whole wetlands (Spieles and Mitsch 2000). In
3457 other cases, both simple and interactive effects have been examined experimentally, e.g., to
3458 separate effects of hydrologic regime from nutrient loading (Neill 1990a, b; Neill 1992, Bayley
3459 et al., 1985). If nutrient effects are examined by comparing condition of natural wetlands along a
97
-------
December 2006 - DRAFT Chapter 7. Data Analysis: Experiments, Observational Data, and Modeling
3460 loading or concentration gradient, effects of other driving factors can be minimized by making
3461 comparisons among wetlands of similar hydrogeomorphic type and climatic regime within a
3462 well-defined sampling window. In addition, multivariate techniques for partitioning effects
3463 among multiple factors can be used, such as partial CCA or partial redundancy analysis (Cooper
3464 et al., 1999, Jensen et al., 1999).
3465
3466
3467 7.6 STATISTICAL TECHNIQUES
3468
3469 Quantitative methods can be used to assess metric cost, evaluation, response variability, and
3470 ability to meet data quality objectives. The most appropriate method varies with respect to the
3471 indicator or variable being considered. In general, statistical techniques are aimed at making
3472 conjectures or inferences about a population's values or relationships between variables in a
3473 sample randomly taken from the population of interest. In these terms, population is defined as
3474 all possible values that a certain parameter may take. For example, in the case of total
3475 phosphorus levels present in marsh sediments in nutrient ecoregion VII, the total population
3476 would be determined if all the marshes in that ecoregion were sampled, which would negate the
3477 need for data analysis. Practically, a sample is taken from the population and the characteristics
3478 associated with that sample (mean, standard deviation) are "transferred" to the entire population.
3479 Many of the basic statistical techniques are designed to quantify the reliability of this transferred
3480 estimate by placing a confidence interval over the sample-derived parameter. More complex
3481 forms of data analysis involve comparisons of these parameters from different populations (for
3482 example, comparison between sites) or the establishment of complex data models that are
3483 thought to better describe the original population structure (for example, regression). They are
3484 still basic inference techniques that utilize sample characteristics to make conjectures about the
3485 original population.
3486
3487 A basic and typical issue facing any type of sampling design is the number of samples that
3488 should be taken to be confident in the translation from samples to population. The degree of
3489 confidence required should be defined as data quality objectives by the end-user and should
3490 identify the expected statistical rigor for those objectives to be met. There are extensive texts on
3491 types and manners of sampling schemes; these will not be discussed here. This section is geared
3492 to determining the minimum data set recommended to work with subsequent sections of the data
3493 analysis chapter. In interpreting the results of various forms of data analysis, an acceptable level
3494 of statistical error is formulated, this is called type I error, or alpha (a). Type I error can be
3495 defined as the probability of rejecting the null hypothesis (H0) when this is actually true. In
3496 setting the type I error rate, the type II error rate is also specified. The type II error rate, or beta
3497 (|3), is defined as failing to reject the null hypothesis when it is actually false, i.e., declaring that
3498 no significant effect exists when in reality this is the case. In setting the type I error rate, an
3499 acceptable level of risk is recommended, the risk of concluding that a significance exists when
3500 this is not the case in reality, i.e., the risk of a "false positive"(type I error) or "false negative"
3501 (type II error). The concepts of Type I and Type II errors are introduced in Chapter 4 with
98
-------
December 2006 - DRAFT Chapter 7. Data Analysis: Experiments, Observational Data, and Modeling
3502 reference to sampling design and monitoring, and more fully discussed in Chapter 8 with
3503 reference to criteria development.
3504
3505 In experimental or sampling design, of greater interest is a statistic associated with beta (P),
3506 specifically 1 - P, which is the power of a statistical tests. Power is the ability of the statistical
3507 test to indicate significance based on the probability that it will reject a false null hypothesis.
3508 Statistical power depends on the level of acceptable statistical significance (usually expressed as
3509 a probability 0.05 - 0.001 (5% -1%) and termed the a level); the level of power dictates the
3510 probability of "success", or identifying the effect. Statistical power is a function of three factors;
3511 effect size, alpha (a) and sample size, the relationship between the three factors being relatively
3512 complex.
§C 1 1
§U
3515 1. Effect size is defined as the actual magnitude of the effect of interest. This could be the
3516 difference between two means, or the actual correlation between the variables. The
3517 relationship between the effect size and power is intuitive; if the effect size is large (for
3518 example, a large difference between means) this results in a concomitantly large power.
3519
3520 2. Alpha is related to power; to achieve a higher level of significance, power decreases if
3521 other factors are kept constant.
3522
3523 3. Sample size. Generally, this is the easiest factor to control. If the two preceding factors
3524 are set, increased sample sizes will always result in a greater power.
3525
3526 As indicated before, the relationship between these three factors is complex and depends on the
3527 nature of the intended statistical analysis. An online guide for selecting appropriate statistical
3528 procedures is available at: http://www.socialresearchmethods.net/. Software packages for
3529 performing power analysis have been reviewed by Thomas and Krebs (1997). Online power
3530 calculations have been made available by several statistical faculty, and are available at these
3531 websites: http://www.math.yorku.ca/SCS/, http://calculators.stat.ucla.edu/powercalc/,
3532 http://www.survevsystem.com/sscalc.htm,
3533 http://www.health.ucalgary.ca/~rollin/stats/ssize/index.html, http://www.stat.ohio-
3534 state.edu/~jch/ssinput.html, http://www.stat.uiowa.edu. Additional websites are also listed in
3535 Chapter 4 that emphasize designs for monitoring with statistical rigor.
3536
3537 Metric response variability can be evaluated by examining the signal to noise ratio (signal:noise)
3538 along a gradient of nutrient concentrations or loading rates (Reddy et al. 1999). The power of
3539 regression analyses can be determined using the power function for a t-test. Optimization of the
3540 design, such as the spacing, number of levels of observations, and replication at each level,
3541 depend on the purpose of the regression analysis (Neter et al. 1983).
3542
3543 MULTIMETRICINDICIES
3544
99
-------
December 2006 - DRAFT Chapter 7. Data Analysis: Experiments, Observational Data, and Modeling
3545 Multimetric indices are valuable for summarizing and communicating results of environmental
3546 assessments and is one approach in developing criteria. Furthermore, preservation of the biotic
3547 integrity of algal assemblages, as well as fish and macroinvertebrate assemblages, may be an
3548 objective for establishing nutrient criteria. Multimetric indices for stream macroinvertebrates and
3549 fish are common (e.g., Kerans and Karr 1994, Barbour et al. 1999), and multimetric indices with
3550 benthic algae have recently been developed and tested on a relatively limited basis (Kentucky
3551 Division of Water 1993; Hill et al. 2000). Efforts are underway to develop multi-metric indices
3552 of biotic integrity for wetlands, and methods modules are available for characterizing wetland
3553 algal, plant, macroinvertebrate, amphibian, and bird communities
3554 (http://www.epa.gov/waterscience/criteria/wetlands/). Methods for multi-metric indices are
3555 well developed for streams, and these methods are readily transferable to wetlands. However,
3556 higher trophic levels do not often directly respond to nutrients, and therefore may not be as
3557 sensitive to relatively small changes in nutrient concentrations as algal assemblages. It is
3558 recommended that relations between biotic integrity of algal or vegetation assemblages and
3559 nutrients be defined and then related to biotic integrity of macroinvertebrate and fish
3560 assemblages in a stepwise, mechanistic fashion. The practitioner should realize however, that
3561 wetlands with a history of high nutrient loadings have often lost the most sensitive species and in
3562 these cases higher trophic level species may prove to be the best indicators of current nutrient
3563 loadings and wetland nutrient condition.
3564
3565 This section provides an overview for developing a multimetric index that will indicate shifts in
3566 primary producers that are associated with trophic status in wetlands. The first step in developing
3567 a multimetric index of trophic status is to select a set of ecological attributes that respond to
3568 human changes in nutrient concentrations or loading. Attributes that respond to an increase in
3569 human disturbance are referred to as metrics. Six to ten metrics should be selected for the index
3570 based on their sensitivity to human activities that increase nutrient availability (loading and
3571 concentrations), their precision, and their transferability among regions and habitat types.
3572 Selected metrics also should respond to the breadth of biological responses to nutrient conditions
3573 (see discussion of metric properties in McCormick and Cairns 1994).
3574
3575 Effects of nutrients on primary producers and effects of primary producers on the biotic integrity
3576 of macroinvertebrates and fish should be characterized to aid in developing nutrient criteria that
3577 will protect designated uses related to aquatic life (e.g., Miltner and Rankin 1998, King and
3578 Richardson 2002).
3579
3580 Another approach for characterizing biotic integrity of assemblages as a function of trophic
3581 status is to calculate the deviation in species composition or growth forms at assessed sites from
3582 composition in the reference condition. Similarity or dissimilarity indices can be used for the
3583 determining the differences in biotic integrity of a wetland in comparison to the reference
3584 condition. Multivariate similarity or dissimilarity indices need to be calculated for multivariate
3585 attributes such as taxonomic composition (Stevenson 1984; Raschke 1993) as defined by relative
100
-------
December 2006 - DRAFT Chapter 7. Data Analysis: Experiments, Observational Data, and Modeling
3586 abundance of different growth forms or species, or species presence/absence. One standard form
3587 of these indices is percent community similarity (PSC, Whittaker 1952):
3588
3589 PSC =Si=i,s min(ai,bi)
3590
3591 Here a; is the percentage of the ith species in sample a, and b; is the percentage of the same ith
3592 species in a subsequent sample, sample b.
3593
3594 A second common community similarity measurement is based on a distance measurement
3595 (which is actually a dissimilarity measurement, rather than similarity measurement, because the
3596 index increases with greater dissimilarity, Stevenson 1984; Pielou 1984). Euclidean distance
3597 (ED) is a standard distance dissimilarity index, where:
3598
3599 ED = V( Si=i(ai-bi)2)
3600
3601 Log-transformation of species relative abundances in these calculations can increase precision of
3602 metrics by reducing variability in the most abundant taxa. However, the practitioner should also
3603 be aware that transformation, while reducing variability, often decreases sensitivity and the
3604 ability to distinguish true fine scale changes in community and species composition.
3605 Theoretically and empirically, we expect to find that multivariate attributes based on taxonomic
3606 composition more precisely and sensitively respond to nutrient conditions than do univariate
3607 attributes, for instance multimetric algal assemblages (see discussions in Stevenson and Pan
3608 (1999)).
3609
3610 To develop the multimetric index, metrics should be selected and their values normalized to a
3611 standard range such that they all increase with trophic status. Criteria for selecting metrics can be
3612 found in McCormick and Cairns (1994) or many other references. Basically, sensitive and
3613 precise metrics should be selected for the multimetric index and selected metrics should
3614 represent a broad range of impacts and perhaps, designated uses. Values can be normalized to a
3615 standard range using many techniques. For example, if 10 metrics are used and the maximum
3616 value of the multimetric index is defined as 100, all ten metrics should be normalized to the
3617 range of 10 so that the sum of all metrics would range between 0 and 100. The multimetric index
3618 is calculated as the sum of all metrics measured in a system. A high value of this multimetric
3619 index of trophic status would indicate high impacts of nutrients and should be a robust (certain
3620 and transferable) and moderately sensitive indicator of nutrient impacts in a stream. A 1-3-5
3621 scaling technique is commonly used with aquatic invertebrates (Barbour et al. 1999; Karr and
3622 Chu 1999) and could be used with a multimetric index of trophic status as well.
3623
3624
3625 7.7 LINKING NUTRIENT AVAILABILITY TO PRIMARY PRODUCER RESPONSE
3626
101
-------
December 2006 - DRAFT Chapter 7. Data Analysis: Experiments, Observational Data, and Modeling
3627 When evaluating the relationships between nutrients and primary producer response within
3628 wetland systems, it is important to first understand which nutrient is limiting. Once the limiting
3629 nutrient is defined, critical nutrient concentrations can be specified and nutrient-response
3630 relationships developed.
3631
3632 DEFINING THE LIMITING NUTRIENT
3633
3634 The first step in identifying nutrient-producer relationships should be to define the limiting
3635 nutrient. Limiting nutrients will control biomass and productivity within a system. However,
3636 non-limiting nutrients may have other impacts, e.g., toxicological effects related to ammonia
3637 concentrations in sediments or effects on competitive interactions which determine vegetation
3638 community composition (Guesewell et al. 2003). A review of fertilization studies indicated that
3639 vegetation N:P mass ratios are a good predictor of the nature of nutrient limitation in wetlands,
3640 with N:P ratios > 16 indicating P limitation at a community level, and N:P ratios < 14 indicative
3641 of N limitation (Koerselman and Meuleman 1996). Guesewell et al. (2003) found that vegetation
3642 N:P ratios were a good predictor of community-level biomass response to fertilization by N or P,
3643 but for individual species, were only predictive of P-limitation and could not distinguish between
3644 N-limitation, co-limitation, or no limitation. Likewise, N, P, and K levels in wet meadow and fen
3645 vegetation were found to be correlated with estimated supply rates or extractable fractions in
3646 soils (Odle Venterink et al. 2002). A survey of literature values of vegetation and soil total N:P
3647 ratios by Bedford et al. (1999) indicated that many temperate North American wetlands are
3648 either P-limited or co-limited by N and P, especially those with organic soils. Only marshes have
3649 N:P ratios in both soils and plants indicative of N limitation, while soils data suggest that most
3650 swamps are also N-limited.
3651
3652 Many experimental procedures are used to determine which nutrient (N, P, or carbon) limits
3653 algal growth. Algal growth potential (AGP) bioassays are very useful for determining the
3654 limiting nutrient (USEPA 1971). Yet, results from such assays usually agree with what would
3655 have been predicted from N:P biomass ratios, and in some cases N:P ratios in the water. Limiting
3656 nutrient-potential biomass relationships from AGP bottle tests are useful in projecting maximum
3657 potential biomass in standing or slow-moving water bodies. However, they are not as useful in
3658 fast-flowing, and/or gravel or cobble bed environments. Also, the AGP bioassay utilizes a single
3659 species which may not be representative of the response of the natural species assemblage.
3660
3661 Limitation may be detected by other means, such as alkaline-phosphatase activity, to determine
3662 if phosphorus is limiting. Alkaline phosphatase is an extracellular enzyme excreted by some
3663 algal species and from roots in some macrophytes in response to P limitation. This enzyme
3664 hydrolyzes phosphate ester bonds, releasing orthophosphate (PO4) from organic phosphorus
3665 compounds (Mullholland et al. 1991). Therefore, the concentration of alkaline phosphatase in the
3666 water can be used to assess the degree of P limitation. Alkaline phosphatase activity, monitored
3667 over time in a waterbody, can be used to assess the influence of P loads on the growth limitation
3668 of algae (Richardson and Qian 1999).
102
-------
December 2006 - DRAFT Chapter 7. Data Analysis: Experiments, Observational Data, and Modeling
3669
3670 There have been no empirical relationships published relating nutrient concentrations or inputs
3671 to wetland chlorophyll a or productivity levels as there have been for streams and lakes. This is
3672 likely due to the large number of factors interacting with nutrients that determine net ecological
3673 effects in wetlands. For example, eutrophication of Great Lakes coastal wetlands and increases
3674 in agricultural area in upstream watersheds have been correlated with decreases in diversity of
3675 submerged aquatic vegetation, yet researchers were unable to uncouple the effects of nutrients
3676 from those of turbidity (Lougheed et al. 2001). Even in experimentally controlled settings, where
3677 it is possible to separate increased suspended solids loadings from nutrient loadings, effects of
3678 nutrients depend heavily on other factors such as periodicity of nutrient additions (pulse vs. press
3679 loadings; Gabor et al. 1994, Murkin et al., 1994, Hann and Goldsborough 1997, McDougal et al.
3680 1997), water regime (Neill 1990a, b; Thormann and Bayley 1997), food web structure
3681 (Goldsborough and Robinson 1996) and time lags (Neill 1990a, b). It is important in
3682 experimental settings to utilize adequate controls for water additions that may accompany
3683 nutrients (Bayley et al. 1985); in empirical comparisons from field data, it may be difficult if not
3684 impossible to separate out these effects. Day et al. (1988) propose a general conceptual model
3685 describing responses of different wetland plant guilds in riverine wetlands based on a
3686 combination of disturbance regime, hydrologic regime, and nutrients. In the latter case, proper
3687 classification of sites based on disturbance and hydrologic regime prior to describing reference
3688 condition, help to adequately separate out nutrient-related effects and to explain differences in
3689 response.
3690
3691 The significance of food web structure in determining nutrient effects does not preclude deriving
3692 predictive nutrient-primary producer relationships, or minimize the importance of describing
3693 significant impacts. However, it does highlight the importance of adequately characterizing the
3694 trophic structure of wetlands prior to comparison, especially the number of trophic levels (e.g.,
3695 presence or absence of planktivorous fish) and examining interactive effects on multiple classes
3696 of primary producers: phytoplankton, epipelon, epiphytic algae, metaphyton, and macrophytes
3697 (Goldsborough and Robinson 1996, McDougal et al., 1997). In some cases, addition of nutrients
3698 may have little or no effect on some components such as benthic algae, but can create significant
3699 shifts in primary productivity among others, such as a loss of macrophytes and associated
3700 epiphytes with an increase in inedible filamentous metaphyton and shading of the water column
3701 (McDougal et al., 1997).
103
-------
December 2006 DRAFT Chapter 8. Criteria Development
3702 Chapter 8 Criteria Development
3703
3704
3705 8.1 INTRODUCTION
3706
3707 This chapter describes recommendations for setting scientifically defensible criteria for nutrients
3708 in wetlands by using data that address causal and biotic response variables. Causal variables
3709 (external nutrient loading, soil extractable P, soil extractable N, total soil N and P, and water
3710 column N and P), and biotic response variables (vegetation N and P, biomass, species
3711 composition, and algal N and P) and the supporting variables (hydrologic condition,
3712 conductivity, soil pH, soil bulk density, particle size distribution, and soil organic matter), as
3713 described in Chapter 5, provide an overview of environmental conditions and nutrient status of
3714 the wetland; these parameters are considered critical to nutrient assessment in wetlands. (See
3715 also Chapter 5). Several recommended approaches that water quality managers can use to derive
3716 numeric criteria in combination with other biological response variables are presented. These
3717 recommended approaches can be used alone, in combination, or may be modified for use by
3718 State/Tribal water quality managers to derive criteria for wetland systems in their State/Tribal
3719 waters. Recommended approaches for numeric nutrient criteria development presented here
3720 include:
3721
3722 • the use of reference conditions to characterize natural or minimally impaired wetland
3723 systems with respect to causal and exposure indicator variables,
3724 • applying predictive relationships to select nutrient concentrations that will protect
3725 wetland structure and/or function, and
3726 • developing criteria from established nutrient exposure-response relationships (as in the
3727 peer-reviewed published literature).
3728
3729 The first approach is based on the assumption that maintaining nutrient levels within the range of
3730 values measured for reference systems will maintain the biological integrity of wetlands. This
3731 presumes that a sufficient number of reference systems can be identified. The second two
3732 approaches are response-based, hence the level of nutrients associated with biological
3733 impairment should be used to identify criteria. Ideally, both kinds of information (background
3734 variability and exposure-response relationships) will be available for criteria development.
3735 Recommendations are also presented for deriving criteria based on the potential for effects to
3736 downstream receiving waters (i.e., the lake, reservoir, stream, or estuary influenced by
3737 wetlands). The chapter concludes with a recommended process for evaluating proposed criteria,
3738 suggestions of how to interpret and apply criteria, considerations for sampling for comparison to
3739 criteria, potential modifications to established criteria, and final implementation of criteria into
3740 water quality standards.
3741
104
-------
December 2006 DRAFT Chapter 8. Criteria Development
3742 The RTAG is composed of State, Tribal, and Regional specialists that will help the Agency and
3743 States/Tribes establish nutrient criteria for adoption into their water quality standards. Expert
3744 evaluations are important throughout the criteria development process. The data upon which
3745 criteria are based and the analyses performed to arrive at criteria should be assessed for veracity
3746 and applicability.
3747
3748 8.2 METHODS FOR DEVELOPING NUTRIENT CRITERIA
3749
3750 The following discussions focus on three general methods that can be used in developing
3751 nutrient criteria. First, identification of reference or control systems for each established wetland
3752 type and class should be based on either best professional judgment (BPJ) or percentile
3753 selections of data plotted as frequency distributions. The second method uses refinement of
3754 classification systems, models, and/or examination of system biological attributes to assess the
3755 relationships among nutrients, vegetation or algae, soil, and other variables. Finally, the third
3756 method identifies published nutrient and vegetation, algal, and soil relationships and values that
3757 may be used (or modified for use) as criteria A weight of evidence approach with multiple
3758 attributes that combines one or more of these three approaches should produce criteria of
3759 greater scientific validity.
3760
3761 USING REFERENCE CONDITION TO ESTABLISH CRITERIA
3762
3763 One approach to consider in setting criteria is the concept of reference condition. This approach
3764 involves using relatively undisturbed wetlands as reference systems to serve as examples for the
3765 natural or least disturbed ecological conditions of a region. Three recommended ways of using
3766 reference condition to establish criteria are:
3767
3768 1. Characterize reference systems for each class within a region using best professional
3769 judgment and use these reference conditions to define criteria.
3770
3771 2. Identify the 75th to 95th percentile of the frequency distribution for a class of reference
3772 wetlands as defined in Chapter 3 and use this percentile to define the criteria.
3773
3774 3. Calculate a 5th to 25th percentile of the frequency distribution of the general population of
3775 a class of wetlands and use the selected percentile to define the criteria.
3776
3777 Defining the nutrient condition of wetlands within classes will allow the manager to identify
3778 protective criteria and determine which systems may benefit from management action. Criteria
3779 that are identified using reference condition approaches may require comparisons to similar
3780 systems in other States or Tribes that share the ecoregion so that criteria can be validated. The
3781 comparison process should also be developed and documented.
3782
105
-------
December 2006 DRAFT Chapter 8. Criteria Development
3783 Reference wetlands should be identified for each class of wetland within a state or tribal
3784 ecoregion and then characterized with respect to external nutrient loading, water column N and
3785 P, biotic response variables (macrophytes, algae, soils) and supporting environmental conditions.
3786 Wetlands classified as reference quality should be verified by comparing the data from the
3787 reference systems to general population data for each wetland class. Reference systems should
3788 be minimally disturbed and should have biotic response values that reflect this condition.
3789
3790 Conditions at reference sites may be characterized using either of two frequency distribution
3791 approaches ( see 2 and 3 above). In both approaches, an optimal reference condition value is
3792 selected from the distribution of an available set of wetland data for a given wetland class. This
3793 approach may be of limited value at this time, because few States or Tribes currently collect
3794 wetland monitoring data. However, as more wetlands are monitored and more data become
3795 available, this approach may become more viable.
3796
3797 In the first frequency distribution approach, a percentile (75th - 95th is recommended) is selected
3798 from the distribution of causal and biotic response variables of reference systems selected a
3799 priori based on very specific criteria (i.e., highest quality or least impacted wetlands for that
3800 wetland class within a region). The values for variables at the selected quartile are used as the
3801 basis for nutrient criteria.
3802
3803 If reference wetlands of a given class are rare within a given region, or if inadequate information
3804 is available to assign wetlands with historic nutrient data as "reference" versus "impacted"
3805 wetlands, another approach may be necessary. The second frequency distribution approach
3806 involves selecting a percentile of (1) all wetland data in the class (reference and non-reference)
3807 or (2) a random sample distribution of all wetland data within a particular class. Due to the
3808 random selection process, a lower percentile should be selected because the sample distribution
3809 is expected to contain some degraded systems. This option is most useful in regions where the
3810 number of legitimate "natural" reference wetlands is usually very small, such as in highly
3811 developed land use areas (e.g., the agricultural lands of the Midwest and the urbanized east or
3812 west coasts). EPA's recommendation in this case is the 5th to 25th percentile depending upon the
3813 number of "natural" reference systems available. If almost all systems are impaired to some
3814 extent, then a lower percentile, generally the 5th percentile is recommended for selection of
3815 reference wetlands.
3816
3817 Both the 75th percentile for the subset of reference systems and the 5th to 25th percentile from a
3818 representative random sample distribution are only recommendations. The actual distribution of
3819 the observations should be the major determinant of the threshold point chosen. For example, a
3820 bi-modal distribution of sediment or water-column nutrients might indicate a natural breakpoint
3821 between reference and enriched systems. To illustrate, Figure 8.1 shows both options and
3822 illustrates the presumption that these two alternative methods should approach a common
3823 reference condition along a continuum of data points. In this illustration, the 75th percentile of
106
-------
December 2006 DRAFT
Chapter 8. Criteria Development
3824
3825
3826
3827
the reference data distribution produces an extractable soil P reference condition that
-th
corresponds to the 25 percentile of the random sample distribution.
75th percentile of
reference population is
the starting point for
Concentration
3=
LU
_a
ro
Effects thresholds
can help justify
criterion value.
Concentration
Nutrient data from reference waters (blue) or
from all waters (gold) similar physical
characteristics.
Paired nutrient and effects data from waters
with similar physical characteristics.
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
Figure 8.1 Use of frequency distributions of nutrient concentration for establishing criteria (left
graphic) and use of effects thresholds with nutrient concentration for establishing criteria (right
graphic).
The choice of a distribution cut-off to define the upper range of reference wetland nutrient levels
is analogous to defining an acceptable level of type I error, the frequency for rejecting wetlands
as members of the "unimpacted" class when in fact they are part of the reference wetland
population (a false designation of impairment). If a distribution cut-off of 25% is chosen, the rate
of falsely designating wetlands as impaired will be higher than if a distribution cutoff of 5% is
chosen; however, the frequency of committing Type II errors (failing to identify
anthropogenically-enriched wetlands) will be lower. As described previously in Chapter 7, there
is a trade-off between Type I and Type II errors. When additional information is available it may
107
-------
December 2006 DRAFT Chapter 8. Criteria Development
3842 be possible to justify a range of values that are representative of least-impaired wetlands that
3843 would reduce Type I errors on a system by system basis.
3844
3845 It is important to understand that any line drawn through the data may have certain ramifications;
3846 wetlands in poor condition (on the right) should be dealt with through restoration. The wetlands
3847 to the left of the line have nutrient conditions that are protective of aquatic life and should be
3848 managed to maintain their nutrient condition, i.e. their nutrient concentrations should remain
3849 stable to be protective of aquatic life. These wetlands should be protected according to the
3850 State's or Tribe's approved antidegradation policy, and through continued monitoring to assure
3851 that future degradation is prevented.
3852
3853 State or Tribal water quality managers also may consider analyzing wetlands data based on
3854 designated use classifications. Using this approach, frequency distributions for specific
3855 designated uses could be examined and criteria proposed based on maintenance of high quality
3856 systems that are representative of each designated use. For example one criterion could be
3857 derived that protects superior quality wetland habitat (SWLH) and a second criterion could be
3858 identified that maintains good quality wetland habitat (function maintained but some loss of
3859 sensitive species (Figure 8.2); see Office of Water tiered aquatic life use training module:
3860 (http://www.epa.gov/waterscience/biocriteria/modules/wetl01-05-alus-monitoring.pdf). This
3861 recommended approach is designated as the Tiered Aquatic Life Use (TALU), and is being
3862 developed by the EPA Office of Water in a more detailed publication. Using this approach, a
3863 criterion range is created and a greater number of wetland systems will likely be considered
3864 protective of the designated use. In this case, emphasis may be shifted from managing wetland
3865 systems based on a central tendency to managing towards more pristine systems associated with
3866 Tiers I and II. This approach also will aid in prioritizing systems for protection and restoration.
3867 Subsequent management efforts using this approach should focus on improving wetland
3868 conditions so that, over time, plots of wetland data shift to the left (i.e., improved nutrient
3869 condition) of their initial position.
3870
3871 In summary, frequency distributions can aid in setting criteria by describing the natural potential
3872 and best attainable conditions (reference conditions). The number of divisions or tiers used has
3873 significant implications with respect to system management. A single criterion may limit the
3874 flexibility to make management decisions about whether wetlands are meeting the applicable
3875 water quality standards, and there may be considerable ramifications resulting from that
3876 decision. If the distribution is divided into three tiers, the majority of wetlands may be protective
3877 of their designated use (assuming that these wetlands do not contribute to downstream
3878 degradation of water quality), which will minimize management requirements. The method that
3879 is used may depend on the goals of the individual State or Tribe. Some may wish to set criteria
3880 that encourage all State/Tribal wetland systems to be preserved or restored to reference
3881 conditions. Other managers may consider additional options, such as developing criteria
3882 specifically to protect the designated uses established for wetlands in their region.
3883
108
-------
December 2006 DRAFT
Chapter 8. Criteria Development
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
APPLYING PREDICTIVE RELATIONSHIPS
Two fundamental reasons are commonly considered for using biological attributes in developing
nutrient criteria. The concepts basically promote the use of biotic responses or biocriteria to
nutrient enrichment, i.e., both rationales support evaluation of physical and chemical conditions
in conjunction with biological parameters when establishing water quality criteria. The first
reason is that the primary goal of environmental assessment and management is to protect and
restore ecosystem services and ecological attributes, which often are closely related to biological
features and functions in ecosystems. Therefore, it is the effects of nutrients on the living
components of ecosystems that should become the critical determinant of nutrient criteria, rather
than
Native or natural condition
1
Natural
=
o
•o
a
o
U
13
u
°5b
_o
"o
3
Degraded
MAINE TALU
Minimal loss of species; some
density changes may occur
Some replacement of
sensitive-rare species; ^
functions fully
maintained
Some sensitive species
maintained; altered
distributions; functions
largely maintained
Tolerant species show
increasing dominance;
sensitive species are rare;
functions altered Severe alteration of
structure and function
Low Increasing Effects of Stressors Hi§n
Figure 8.2. Tiered Aquatic Life Use model used in Maine.
the actual nutrient concentrations. The second reason for using biocriteria is that attributes of
biological assemblages usually vary less in space and time than most physical and chemical
characteristics measured in environmental assessments. Thus, fewer mistakes in assessment may
occur if biocriteria are employed in addition to physical and chemical criteria. In those
environments where biological attributes change fairly rapidly, such as in Louisiana's coastal
wetland environment where salinity can vary dramatically in response to wet versus drought
109
-------
December 2006 DRAFT Chapter 8. Criteria Development
3908 years, other techniques will need to be developed. Information on some other techniques can be
3909 found at: Louisiana State University's School of the Coast and Environment
3910 rhttp://www.wetlandbiogeochemistry.lsu.edu/ ] and also in interagency efforts through the LA
3911 Dept. of Natural Resources) to assess coastal area ecology.
3912 [http: //data. 1 ca. gov/Ivan6/app/app_c_ch9. pdf ]
3913
3914 Multimetric indices are a special form of indicators of biological condition in which several
3915 metrics are used to summarize and communicate in a single number the state of a complex
3916 ecological system. Multimetric indices for macroinvertebrates and fish are used successfully to
3917 establish biocriteria for aquatic systems in many States, and several States are developing
3918 multimetric indices for wetlands (see http://www.epa.gov/owow website).
3919
3920 Another recommended approach is to identify threshold or non-linear biotic responses to nutrient
3921 enrichment. Some biological attributes respond linearly with increasing nutrient concentrations
3922 whereas some attributes change in a non-linear manner. Non-linear changes in metrics indicate
3923 thresholds along environmental gradients where small changes in environmental conditions
3924 cause relatively great changes in a biological attribute. In an example from the Everglades, a
3925 specific level of P concentration and loadings was associated with a dramatic shift in algal
3926 composition and loss of the calcareous algal mats typical of this system (Figure 8.3). Overall,
3927 metrics or indices that change linearly (typically higher-level community attributes such as
3928 diversity or a multimetric index) provide better variables for establishing biocriteria because they
3929 respond to environmental change along the entire gradient of human disturbance. However,
3930 metrics that change in a non-linear manner along environmental gradients are valuable for
3931 determining where along the environmental gradient the physical and chemical criteria should be
3932 set and, correspondingly, how to interpret other biotic response variables of interest (Stevenson
3933 et al. 2004a).
3934
3935 USING DATA PUBLISHED IN THE LITERATURE
3936
3937 Values from the published literature may be used to develop nutrient criteria if a strong rationale
3938 is presented that demonstrates the suitability of these data to the wetland of interest (i.e., the
3939 system of interest should share the same characteristics with the systems used to derive the
3940 published values). Published data, if there is enough of it, could be used to develop criteria for
3941 (1) reference condition, (2) predictive (cause and effect) relationships between nutrients and
3942 biotic response variables, (3) tiered criteria or (4) criteria that exhibit a threshold response to
3943 nutrients. However, published data from similar wetlands should not substitute for collection and
3944 analysis of data from the wetland or wetlands of interest.
3945
110
-------
December 2006 DRAFT
Chapter 8. Criteria Development
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
w
I
I
90
SO
70
60
50
40
'in
JLr
20
10
n
y*'*"*' k •.•
^ :*• •»»-;¥*;•»••• * .
^ * •
*
6 8 10 12 14
Distance from Point Source (km)
Figure 8.3. Percent calcareous algal mat cover in relation to distance from the P source showing
the loss of the calcareous algal mat in those sites closer to the source (Stevenson et al. 2002).
CONSIDERATIONS FOR DOWNSTREAM RECEIVING WATERS
More stringent nutrient criteria may be appropriate for wetlands that drain into lentic or standing
waters. For example, it is proposed that 35 ug/L TP concentration and a mean concentration of 8
ug/L chlorophyll a constitute the dividing line between eutrophic and mesotrophic lakes (OECD
1982). Natural nutrient concentrations in some wetlands may be higher than downstream lakes.
In addition, assimilative capacity for nutrients without changes in valued attributes may also be
higher in wetlands than lakes. Nutrient criteria for wetlands draining into lakes should protect
the downstream waters of receiving lakes in addition to wetlands. Therefore, nutrient criteria for
wetlands draining into lakes may need to be lower than typically would be set if only effects on
wetlands were considered.
8.3 EVALUATION OF PROPOSED CRITERIA
Following criteria derivation, an expert assessment of the proposed criteria and their
applicability to all wetlands within the class of interest is encouraged. Criteria should be verified
in many cases by comparing criteria values for a wetland class within an ecoregion across State
and Tribal boundaries. In fact, development of interstate criteria should be an integral part of a
State or Tribe's water quality standards program. In addition, prior to recommending any
proposed criterion, it is recommended that States and Tribes take into consideration the water
quality standards of downstream waters to ensure that their water quality standards provide for
ill
-------
December 2006 DRAFT Chapter 8. Criteria Development
3972 attainment and maintenance of the water quality standards of downstream waters, (see 40 C.F.R.
3973 131.10(b)). Load estimating models, such as those recommended by EPA (USEPA 1999), can
3974 assist in this determination (see External Nutrient Loading in Chapter 5.3). Water quality
3975 managers responsible for downstream receiving waters also should be consulted.
3976
3977
3978 8.4 INTERPRETING AND APPLYING CRITERIA
3979
3980 After evaluating criteria proposed for each wetland class, determining wetland condition in
3981 comparison with nutrient criteria can be made by following these steps:
3982
3983 1. Calculate duration and frequency of criteria exceedences as well as associated
3984 consequences. This can be done using modeling techniques or correlational analysis of
3985 existing data.
3986
3987 2. Develop and test hypotheses to determine agreement with criteria. Analyze for alpha
3988 (Type I) and beta (Type II) errors (see Chapter 7).
3989
3990 3. Reaffirm appropriateness of criteria for protecting designated uses and meeting water
3991 quality standards (i.e., by effective sampling and monitoring of the wetlands).
3992
3993 The goal is to identify highly protective criteria and standards. Criteria should be based on
3994 ecologically significant changes as well as statistically significant differences in compiled data.
3995 Although criteria are developed exclusively based on scientifically defensible methods, adoption
3996 of water quality standards also allows consideration of social, political, and economic factors.
3997 Thus, it is imperative that some determination is given during the criteria development process to
3998 how criteria can be implemented into standards that are defensible to the public and regulated
3999 communities, and effectively translated into permits, TMDLs, or watershed implementation
4000 plans for nonpoint source nutrient management.
4001
4002
4003 8.5 SAMPLING FOR COMPARISON TO CRITERIA
4004
4005 Sampling to evaluate agreement with the standards implemented from nutrient criteria should be
4006 carefully defined to ensure that state or tribal sampling is compatible with the procedures used to
4007 establish the criteria. If State or Tribal observations are averaged over the year, balanced
4008 sampling is essential and the average should not exceed the criterion. In addition, no more than
4009 ten percent of the observations contributing to that average value should exceed the criterion.
4010
112
-------
December 2006 DRAFT Chapter 8. Criteria Development
4011 It is important to note that, in some regions where nutrient impacts occur seasonally depending
4012 on precipitation and temperature regimes, sampling and assessment should focus on the period
4013 (e.g.,index period) when impacts are most likely to occur.
4014
4015 A load estimating model may be applied to a watershed to back-calculate the criteria
4016 concentration for an individual wetland from its load allocation. This approach to criteria
4017 determination also may be applied on a seasonal basis and should help States/Tribes relate their
4018 wetland criteria with their stream, lake, or estuarine criteria. It also may be particularly important
4019 for criteria developed for wetlands that cross State/Tribal boundaries.
4020
4021
4022 8.6 CRITERIA MODIFICATIONS
4023
4024 There may be specific cases identified by States or Tribes that require modification of
4025 established criteria, either due to unique wetland system characteristics or specific designated
4026 uses approved for a wetland. Two examples of acceptable criteria modifications are presented
4027 below.
4028
4029 SITE SPECIFIC CRITERIA
4030
4031 If a State or Tribe has additional information and data that indicate a different value or set of
4032 values is more appropriate for specific wetland systems than ecoregionally-derived criteria, the
4033 State or Tribe may decide to develop site-specific criteria modifications. This value can be
4034 incorporated into State or Tribal water quality standards and submitted to EPA for approval.
4035
4036 DESIGNATED USE APPROACHES
4037
4038 Once a regional criterion has been established, it should be reviewed and calibrated periodically.
4039 Any State or Tribe in the region with similar classes of wetlands may elect to use the criterion as
4040 the basis for developing its own criteria to protect its designated uses for specific wetland
4041 classes. This is entirely appropriate as EPA expects criteria developed using one of the
4042 approaches recommended here will be protective of aquatic life in wetlands and scientifically
4043 defensible.
113
-------
December 2006- DRAFT REFERENCES
4044
4045 REFERENCES
4046
4047
4048 Adamus P. R. 1992. Choices in Monitoring Wetlands. In: Ecological Indicators. DH McKenzie,
4049 DE Hyatt, VJ McDonald (eds).New York: Elsevier Applied Science, pp. 571-592.
4050
4051 Allen S. E, H. M. Grimshaw, and A. P. Rowland. 1986. Chemical analysis. In: Methods in Plant
4052 Ecology. Moore P. D., S. B. Chapman (eds.).Boston: Blackwell Scientific Publications,
4053 pp. 285-344.
4054
4055 Amoozegar, A. and A. W. Warrick. 1986. Hydraulic conductivity of saturated soils: field
4056 methods, pp. 735-770. In: Methods of Soil Analysis, Part 1, Physical and Mineralogical
4057 Methods. Agronomy monograph No. 9. A. Klute (ed.) American Society of
4058 Agronomy-Soil Science Society of America, Madison WI.
4059
4060 Anderson, J. M. 1976. An ignition method for determination of total phosphorus in lake
4061 sediments. Water Research. 10:329-331.
4062
4063 Angeler, D. G., P. Chow-Fraser, M. A. Hanson, S. Sanchez-Carrille, and K. D. Zimmer. 2003.
4064 Biomanipulation: a useful tool for freshwater wetland mitigation? Freshwater Biology
4065 48:2203-2213.
4066
4067 Apfelbeck, R. S. 1999. Development of biocriteria for wetlands in Montana. Montana Dept. of
4068 Environmental Quality, Helena, MT.
4069
4070 APHA. 1999. Standard Methods for Examination of Water and Wastewater. 21st ed. Eaton, A.
4071 D., L. C. Clesceri, and A. E. Greenberg (eds.). American Public Health Association,
4072 Washington, DC.
4073
4074 Arnold, J. G., J. R. Williams, R. Srinivasan, K. W. King, and R. H. Griggs. 1995. SWAT- Soil
4075 and Water Assessment Tool: Draft Users Manual. USDA-ARS, Temple, TX.
4076
4077 Bailey, R. G. 1976. Ecoregions of the United States (map). Ogden, UT: U.S. Department of
4078 Agriculture, Forest Service. Intermountain Region. Scale 1:7,500,000.
4079
4080 Bailey, S. E; J. Jr. Zoltek, A. J. Hermann, T. J. Dolan, and L. Tortora. 1985. Experimental
4081 manipulation of nutrients and water in a freshwater marsh: Effects on biomass,
4082 decomposition, and nutrient accumulation. Limnology and Oceanography 30: 500-512.
4083
4084 Ball, J. E., M. J. White, G. de R. Innes, and L. Chen. 1993. Application of HSPF on the Upper
4085 Nepean Catchment. In: pp. 343-348. Proceedings of Hydrology and Water Resources
4086 Symposium. Newcastle, New South Wales, Australia.
114
-------
December 2006- DRAFT REFERENCES
4087
4088 Balla, S. A. and J. A. Davis. 1995. Seasonal variation in the macroinvertebrate fauna of wetlands
4089 of differing water regime and nutrient status on the Swan coastal plain, Western
4090 Australia. Hydrobiologia 299:147-161.
4091
4092 Barbour, M. T., J. Gerritsen, B. D. Snyder, and J. B. Stribling. 1999. RapidBioassessment
4093 Protocols for Use in Wadeable Streams and Rivers: Periphyton, Benthic
4094 Macroinvertebrates, andFish 2nd ed. U.S. Environmental Protection Agency, Office of
4095 Water, Washington, DC. EPA 841-B-99-002.
4096
4097 Barko, J. W. 1983. The growth of Myriophyllum spicatum in relation to selected characteristics
4098 of sediment and solution. Aquatic Botany 15:91-103.
4099
4100 Barko, J. W. and R. M. Smart. 1986. Sediment-related Mechanisms of Growth Limitation in
4101 Submersed Macrophytes. J. Ecol. 67(5): 1340.
4102
4103 Bayley, S. E. 1985. Effect of Natural Hydroperiod Fluctuations on Freshwater Wetlands
4104 Receiving Added Nutrients. In: Ecological Considerations in Wetlands Treatment of
4105 Municipal Wastewaters. Van Nostrand Reinhold Company New York. pp. 180-9
4106
4107 Bayley, S. E., J. Zoltek, Jr., A. J. Hermann, T. J. Dolan, and L. Tortora. 1985. Experimental
4108 manipulation of nutrients and water in a freshwater marsh: Effects on biomass,
4109 decomposition, and nutrient accumulation. Limnology and Oceanography 30:500-512.
4110
4111 Beals, E. W. 1984. Bray-Curtis ordination: an effective strategy for analysis of multivariate
4112 ecological data. Advances in Ecological Research 14:1-55.
4113
4114 Bedford, B. L., M. R. Walbridge, and A. Aldous. 1999. Patterns in nutrient availability and plant
4115 diversity of temperate North American wetlands. Ecology 80:2151-69.
4116
4117 Bennett, R. J. 1999. Examination of macroinvertebrate communities and development of an
4118 invertebrate community index (ICI) for central Pennsylvania wetlands. (M.S. thesis,
4119 Pennsylvania State University).
4120
4121 Bicknell, B. R., J. C. Imhoff, J. L. Kittle, A. S. Donigian, and R. C. Johanson, 1993.
4122 Hydrological Simulation Program - FORTRAN (HSPF): User's manual for release 10.0.
4123 Environmental Research Laboratory, U.S. Environmental Protection Agency, Athens,
4124 GA. PA 600/3-84-066.
4125
4126 Blake, G.R., and K.H. Hartge. 1986. Bulk Density. In: A. Methods of Soil Analysis, Parti.
4127 Physical andMineralogicalMethods: Agronomy Monograph no. 9 (2nd ed.), Klute, ed.
4128 pp. 363-375.
4129
115
-------
December 2006- DRAFT REFERENCES
4130 Bollens, U. and D. Ramseier. 2001. Shifts in abundance of fen-meadow species along a nutrient
4131 gradient in a field experiment. Bulletin of the Geobotanical Institute 67:57-71.
4132
4133 Boyer, K. E. and P. Fong. 2005. Macroalgal-mediated transfers of water column nitrogen to
4134 intertidal sediments and salt marsh plants. Journal of Experimental Marine Biology and
4135 Ecology 321(l):59-69.
4136
4137 Bremner, J. M. 1996. Nitrogen- Total. In: Methods of Soil Analysis. Part 3, Chemical Methods B
4138 SSSA Book Series No. 5. D. L. Sparks et al. (eds.). Soil Science Society of America Inc.,
4139 pp. 1085-1121.
4140
4141 Bridgham S. D., and C. J. Richardson. 1993. Hydrology and nutrient gradients in North Carolina
4142 peatlands. Wetlands 13:207-218.
4143
4144 Bridgham, S. D., K. Updegraff, and J. Pastor. 1998. Carbon, nitrogen, and phosphorus
4145 mineralization in northern wetlands. Ecology 79: 1545-1561.
4146
4147 Brinson, M. M. 1993. A Hydrogeomorphic Classification for Wetlands. U.S. Army Corps of
4148 Engineers, Waterways Experiment Station. Washington, DC. Wetlands Research
4149 Program Technical Report WRP-DE-4.
4150
4151 Brinson M. M., H. D. Bradshaw, E. S. Kane 1984. Nutrient assimilative capacity of an alluvial
4152 floodplain swamp. J Appl. Ecol. 21:1041-1057.
4153
4154 Brinson, M. M., R. D. Rheinhardt, F. R. Hauer, L. C. Lee, W. L. Nutter, R. D. Smith, and D.
4155 Whigham. 1995. A guidebook for application of hydrogeomorphic asessments to riverine
4156 wetlands. U.S. Army Corps of Engineers, Washington, DC. Wetlands Research Program
4157 Technical Report WRP-DE-11.
4158
4159 Broome S. W., E. D. Seneca, and W. W. Jr. Woodhouse. 1983. The effects of source, rate and
4160 placement of nitrogen and phosphorus fertilizers on growth of Spartina alterniflora
4161 transplants in North Carolina. Estuaries 6:212-226.
4162
4163 Broome S. W., E. D. Seneca, and W. W., Jr. Woodhouse. 1986. Long-term growth and
4164 development of transplants of the salt-marsh grass Spartina alterniflora. Estuaries
4165 9:63-74.
4166
4167 Brown, M, University of Florida, Center for Wetlands, personal communication.
4168
4169 Brown, M. T. and M. B. Vivas. 2004. A landscape development intensity index. Environmental
4170 Monitoring and Assessment 101: 289-309.
4171
116
-------
December 2006- DRAFT REFERENCES
4172 Busnardo, M. I, R. M. Gersberg, R. Langis, T. L. Sinicrope, and J. B. Zedler. 1992. Nitrogen
4173 and phosphorus removal by wetland mesocosms subjected to different hydroperiods.
4174 Ecological Engineering 1: 287-307.
4175
4176 Carlson R. E. 1977. A tropic state index for lakes. Limn. Oceanogr. 22:361-369.
4177
4178 Carpenter, S.R., N.F. Caraco, D.L. Cornell, R.W. Howarth, A.N. Sharpley and V.H. Smith. 1998.
4179 Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecol Appl 8:559-
4180 568.
4181
4182 Chalmers A. G. 1979. The effects of fertilization on nitrogen distribution in a Spartina
4183 alterniflora salt marsh. Estuarine Coastal Marine Sci. 8:327-337.
4184
4185 Chessman, B. C., K. M. Trayler, and J. A. Davis. 2002. Family- and species-level biotic indices
4186 for macroinvertebrates of wetlands on the Swan Coastal Plain, Australia. Marine and
4187 Freshwater Research 53:919-30.
4188
4189 Chiang C., C. B. Craft, D. W. Rogers, and C. J. Richardson. 2000. Effects of four years of
4190 nitrogen and phosphorus additions on Everglades plant communities. Aqua Bot 68:61-78.
4191
4192 Clairain, EJ. (2002). "Hydrogeomorphic Approach to Assessing Wetland Functions: Guidelines
4193 for Developing Regional Guidebooks; Chapter 1, Introduction and Overview of the
4194 Hydrogeomorphic Approach," ERDC/EL TR-02-3, U.S. Army Engineer Research and
4195 Development Center, Vicksburg, MS.
4196
4197 Cohn, T.A., DeLong, L.L., Gilroy, E.J., Hirsch, R.M., and Wells, O.K., 1989, Estimating
4198 constituent loads: Water Resources Research, v. 25, no. 5, p. 937-942.
4199
4200 Cole, A. C., R. P. Brooks, and D. H. Wardrop. 1997. Wetland hydrology as a function of
4201 hydrogeomorphic (HGM) subclass. Wetlands 17:456-467.
4202
4203 Cooper, S. R., J. Huvane, J. P. Vaithiyanathan, and C. J. Richardson. 1999. Calibration of
4204 diatoms along a nutrient gradient in Florida Everglades Water Conservation Area-2A,
4205 USA. Journal of Paleolimnology 22: 413-437.
4206
4207 Cowardin, L .M., V. Carter, F. C. Golet, and E. T. LaRoe. 1979. Classification of wetlands and
4208 deepwater habitats of the United States. U.S. Fish & Wildlife Service Pub. FWS/OBS-
4209 79/31, Washington, D. C.
4210
4211 Craft, C.B. 2001. Soil organic carbon, nitrogen and phosphorus as indicators of recovery in
4212 restored Spartina marshes. Ecological Restoration 19:87-91.
4213
117
-------
December 2006- DRAFT REFERENCES
4214 Craft, C.B., J.P. Megonigal, S.W. Broome, J. Cornell, R. Freese, R.J Stevenson, L. Zheng and J.
4215 Sacco. 2003. The pace of ecosystem development of constructed Spartina alterniflora
4216 marshes. Ecological Applications 13:1317-1432.
4217
4218 Craft C. B. and C. J. Richardson. 1993a. Peat accretion and phosphorus accumulation along a
4219 eutrophication gradient in the Northern Everglades. Biogeochem. 22:133-156.
4220
4221 Craft, C.B. and CJ. Richardson. 1993b. Peat accretion and N, P, and organic C accumulation in
4222 nutrient-enriched and unenriched Everglades peatlands. Ecol. Appl. 3 (3): 446-458.
4223
4224 Craft C. B. and C. J. Richardson. 1998. Recent and long-term organic soil accretion and nutrient
4225 accumulation in the Everglades. Soil Science Society of America Journal 62: 834-843
4226
4227 Craft C. B., Vymazal J., and C. J. Richardson 1995. Response of Everglades plant communities
4228 to nitrogen and phosphorus additions. Wetlands 15:258-271.
4229
4230 D'Angelo, E.M. and K.R. Reddy. 1994a. Diagenesis of organic matter in a wetland receiving
4231 hypereutrophic lake water. I. Distribution of dissolved nutrients in the soil and water-
4232 column. Journal of Environmental Quality 23:937-943.
4233
4234 D'Angelo, E.M. and K.R. Reddy. 1994b. Diagenesis of organic matter in a wetland receiving
4235 hypereutrophic lake water. II. Role of inorganic electron acceptors in nutrient release.
4236 Journal of Environmental Quality 23:928-936.
4237
4238 Davies, S. P., and S. K. Jackson. 2006. The biological condition gradient: a descriptive model for
4239 interpreting change in aquatic ecosystems. Ecological Applications 16:1251-1266.
4240
4241 Davis, C. B. and A. G. van der Valk. 1983. Uptake and release of nutrients by living and
4242 decomposing Typha glauca Godr. tissues at Eagle Lake, Iowa. AquaticBotany 16:75-89.
4243
4244 Davis S. M. 1991. Growth, decomposition and nutrient retention of Cladium jamaicense Crantz
4245 and Typha domingensis Pers. in the Florida Everglades. Aqua Bot 40:203-224.
4246
4247 Davis, S. and J. Ogden. 1994. Everglades: The Ecosystem and Its Restoration. St. Lucie Press,
4248 Delray Beach, FL. 848 pp.
4249
4250 Day, R. T., P. A. Keddy, J. McNeill, and T. Carleton. 1988. Fertility and disturbance gradients:
4251 A summary model for riverine marsh vegetation. Ecology 69(4): 1044-54.
4252
4253 DeBusk, W. F., K. R. Reddy, M. S. Koch, and Y. Wang. 1994. Spatial distribution of soil
4254 nutrients in a northern Everglades marsh: Water Conservation Area 2A. Soil Sci. Soc.
4255 Am. J. 58:543-552.
118
-------
December 2006- DRAFT REFERENCES
4256
4257 Detenbeck N. E., D. L. Taylor, and A. Lima. 1996. Spatial and temporal variability in wetland
4258 water quality in the Minneapolis/St. Paul, MN, metropolitan area. Environ Monitor
4259 Assess 40:11-40.
4260
4261 Detenbeck, N. E., S. L. Batterman, V. J. Brady, J. C. Brazner, V. M. Snarski, D. L. Taylor, J. A.
4262 Thompson, and J. W. Arthur. 2000. A test of watershed classification systems for
4263 ecological risk assessment. Environmental Toxicology and Chemistry 19(4):1174-81.
4264
4265 Dodds, W. K., V. H. Smith, and B. Zander. 1997. Developing nutrient targets to control benthic
4266 chlorophyll levels in streams: A Case Study of the Clark Fork River. Water Res.
4267 31:1738-1750.
4268
4269 Donigian, A. S., Jr., B. R. Bicknell, L. C. Linker, J. Hannawald, C. Chang, and R. Reynolds.
4270 1990. Chesapeake Bay Program Watershed Model Application to Calculate Bay Nutrient
4271 Loadings: Preliminary Phase I Findings and Recommendations. Prepared by Aqua Terra
4272 Consultants for U.S. EPA Chesapeake Bay Program, Annapolis, MD.
4273
4274 Donigian, A. S. Jr., B. R. Bicknell, and J. C. Imhoff 1995. Hydrological Simulation Program -
4275 FORTRAN (HSPF). In: Computer Models of Water shed Hydrology, V.P. Singh, ed.,
4276 Water Resources Publications, Littleton, CO. pp. 395-442.
4277
4278 Dougherty, S. J, C. R. Lane, and M. T. Brown. 2000. Proposed classification for biological
4279 assessment of Florida inland freshwater wetlands. 33 pp.
4280
4281 Ehrenfeld, J. G. and J. P. Schneider. 1993. Responses of forested wetland vegetation to
4282 perturbations of water chemistry and hydrology. Wetlands 13:122-129.
4283
4284 Eilers, J. M., G. E. Glass, K. E. Webster, and J. A. Rogalla. 1983. Hydrologic control of lake
4285 susceptibility to acidification. Canadian Journal of Fisheries and Aquatic Sciences
4286 40:1896-1904.
4287
4288 ELI (Environmental Law Institute). 1988. Almanac of Enforceable State Laws to Control
4289 Nonpoint Source Water Pollution. Washington, DC. ELI Project #970301.
4290
4291 Ellison, A. M. and B. L. Bedford. 1995. Response of a wetland vascular plant community to
4292 disturbance: A simulation study. Ecological Applications 5: 109-123.
4293
4294 Ernst, T. L., N. C. Leibowitz, D. Roose, S. Stehman, and N. S. Urquhart. 1995. Evaluation of
4295 USEPA Environmental Monitoring and Assessment Program's (EMAP)-Wetlands
4296 sampling design and classification. Environmental Management 19:99-113.
4297
119
-------
December 2006- DRAFT REFERENCES
4298 Ewel, K. C. 1976. Seasonal changes in the distribution of water fern and duckweed in cypress
4299 domes receiving sewage. Third Annual Report on Cypress Wetlands, Florida University,
4300 Center for Wetlands, pp. 164-70.
4301
4302 Finlayson, M., P. Cullen, D. Mitchell, and A. Chick. 1986. An assessment of a natural wetland
4303 receivig sewage effluent. Australian Journal of Ecology 11:33-47.
4304
4305 Fong, P., Boyer, K. E. and J. B. Zedler. 1990. Developing an indicator of nutrient enrichment in
4306 coastal estuaries and lagoons using tissue nitrogen content of the opportunistic alga,
4307 Enteromorpha intestinalis (L. Link). Journal of Experimental Marine Biology and Ecology
4308 231(l):63-79.
4309
4310 Fry, B. 1984. 13C/12C ratios and the trophic importance of algae in Florida Syringodium filiforme
4311 seagrass meadows. Mar. Biol. 79:11-19.
4312
4313 Gabor, T. S, H. R. Murkin, M. P. Stainton, J. A. Boughen, and R. D. Titman. 1994. Nutrient
4314 additions to wetlands in the Interlake region of Manitoba, Canada: Effects of a single
4315 pulse addition in spring. Hydrobiologia 279-280: 497-510.
4316
4317 Galatowitsch, S. M. and A. G. van der Valk. 1996. The vegetation of restored and natural prairie
4318 wetlands. Ecological Applications 6:102-112.
4319 Gauch, H. G., Jr. 1982. Noise reduction by Eigenvector ordinations. Ecology 63:1643-49.
4320
4321 Gaudet, C. L. and P. A. Keddy. 1995. Competitive performance and species distribution in
4322 shoreline plant communities: a comparative approach. Ecology 76: 280-91.
4323
4324 Gephardt, K., S. Leonard, G. Staidl, and D. Prichard. 1990. Riparian land management: riparian
4325 and wetland classification review. Bureau of Land Management, Lakewood, CO. Tech.
4326 Paper TR-1737-5.
4327
4328 Gerloff, G. C. and P. H. Krombholz. 1966. Tissue analysis as a measure of nutrient availability
4329 for the growth of angiosperm aquatic plants. Limnol. Oceanogr. 11: 529-537.
4330
4331 Goldsborough, L. G. and G. G. C. Robinson. 1996. Pattern in wetlands, Chapter 4. In: Algal
4332 Ecology in Freshwater Benthic Ecosystems. R. J. Stevenson, M. L. Bothwell, R. L. Lowe,
4333 eds., Academic Press, pp. 77-117.
4334
4335 Golet, F. C. and J. S. Larson. 1974. Classification of freshwater wetlands in the glaciated
4336 Northeast. U.S. Fish and Wildlife Service Resources Publ. 116, Washington, D.C.
4337
4338 Gorham, E., W. E. Dean, and J. E. Sager. 1983. The chemical composition of
4339 lakes in the north-central United States. Limnology and Oceanography
120
-------
December 2006- DRAFT REFERENCES
4340 28:287-301.
4341
4342 Green R. H. 1979. Sampling Design and Statistical Methods for Environmental Biologists. New
4343 York: Wiley.
4344
4345 Green, E.K. and S.M. Galatowitsch. 2002. Effects ofPhalaris arundinacea and nitrate-N
4346 addition on the establishment of wetland plant communities. Journal of Applied
4347 Ecology 39 (1), 134-144.
4348
4349 Groffman P. M. 1994. Denitrification in freshwater wetlands. Current Topics Wetland
4350 Biogeochem. 1:15-35.
4351
4352 Grossman, D. H., D. Faber-Langendoen, A. S. Weakley, M. Anderson, P. Bourgeron, R.
4353 Crawford, K. Goodin, S. Landaal, K. Metzler, K. Patterson, M. Pyne, M. Reid, and L.
4354 Sneddon. 1998. International Classification of Ecological Communities: Terrestrial
4355 Vegetation of the United States. Volume I. The National Vegetation Classification
4356 System: Development, Status, and Applications. The Nature Conservancy, Arlington,
4357 VA, USA.
4358
4359 Growns, J. E., J. A. Davis, F. Cheal, L. G. Schmidt, R. S. Rosich. 1992. Multivariate Pattern
4360 Analysis of Wetland Invertebrate Communities and Environmental Variables in Western
4361 Australia. Australian Journal of Ecology 17: 275-288.
4362
4363 Guesewell, S., W. Koerselman, and J. T. A. Verhoeven. 1998. The N:P ratio and the nutrient
4364 limitation of wetland plants. Bulletin of the Geobotanical Institute 64:77-90.
4365
4366 Guesewell, S., W. Koerselman, and J. T. A. Verhoeven. 2003. N:P ratios as indicators of nutrient
4367 limitation for plant populations in wetlands. Ecological Applications. 13: 372-384.
4368
4369 Guntenspergen, G. R., W. Kappel, and F. Stearns. 1980. Response of a bog to application of
4370 lagoon sewage: The Drummond Project- an operational trial. In: Proceedings of the 6th
4371 International Peat Congress. Duluth, MN, USA.
4372
4373 Gwin S. E., M. E. Kentula, and P. W. Shaffer. 1999. Evaluating the effects of wetland regulation
4374 through hydrogeomorphic classification and landscape profiles. Wetlands 19(3):477-489.
4375
4376 Haith, D. A. and L. L. Shoemaker. 1987. Generalized watershed loading functions for stream
4377 flow nutrients. Water Resources Bulletin 107:121-137.
4378
4379 Haith, D. A., R. Mandel, and R. S. Wu. 1992. GWLF - Generalized Water shed Loading
4380 Functions, Version 2.0 - User's manual. Department of Agricultural Engineering, Cornell
4381 University, Ithaca, NY.
121
-------
December 2006- DRAFT REFERENCES
4382
4383 Halsey L., D. Vitt, and S. Zoltai. 1997. Climatic and physiographic controls on wetland type and
4384 distribution in Manitoba, Canada. Wetlands 17:243-262.
4385
4386 Hamilton, S. K., W. M. Lewis, and S. J. Sippel. 1992. Energy sources for aquatic animals in the
4387 Orinoco River floodplain - evidence from stable isotopes. Oecologia. 89:324-330.
4388
4389 Hann, B. J. and L. G. Goldsborough. 1997. Responses of a prairie wetland to press and pulse
4390 additions of inorganic nitrogen and phosphorus: invertebrate community structure and
4391 interactions. Archiv fuer Hydrobiologie 140:169-94.
4392
4393 Hann, B., C. Mundy, and L. Goldsborough. 2001. Snail-periphyton interactions in a prairie
4394 lacustrine wetland. Hydrobiologia 457:167-75.
4395
4396 Harris S. C., T. H. Martin, and K. W. Cummins. 1995. A model for aquatic invertebrate response
4397 to Kissimmee River restoration. Restoration Ecol 3:181-194.
4398
4399 Hayek L. C. 1994. Research Design for Quantitative Amphibian Studies. In: Measuring and
4400 Monitoring Biological Diversity, Standard Methods for Amphibians. Washington, DC:
4401 Smithsonian Institution Press.
4402
4403 Heikkila, H. 1987. Vegetation and Ecology of Mesotrophic and Eutrophic Fens in Western
4404 Finland. Annales Botanici Fennici 24: 155-175.
4405
4406 Herdendorf, C. E., S. M. Hartley, and M. D. Barnes. 1981. Fish and wildlife resources of the
4407 Great Lakes coastal wetlands. Vol 1: Overview. U.S. Fish and Wildlife Service
4408 FWS/OBS-81/02-Vl.pp. 469.
4409
4410 Hill, B. H., A. T. Herlihy, P. R. Kaufmann, R. J. Stevenson, F. H. McCormick, and C. B.
4411 Johnson. 2000. The use of periphyton assemblage data as an index of biotic integrity. J.
4412 N. Amer. Benthol. Soc. 19:50-67.
4413
4414 Hopkinson, C. S. and J. P. Schubauer. 1980. Static and dynamic aspects of nitrogen cycling in
4415 the salt marsh graminoid Spartina alterniflora. Ecology 65 (3), pp. 961-969.
4416
4417 Hughes, R. M. and J. M. Omernik. 1981. A proposed approach to determine regional patterns in
4418 aquatic ecosystems. In: Acquisition and utilization of aquatic habitat inventory
4419 information. Proceedings of a symposium. October 28-30, 1981. pp. 92-102. Portland,
4420 OR: Western Division, American Fisheries Society.
4421
122
-------
December 2006- DRAFT REFERENCES
4422 Hurlbert, S. H. 1984. Pseudoreplication and the design of ecological field experiments. Ecol
4423 Moog 54(2): 187-211.
4424
4425 Jackson, L. E., J. C. Kurtz, and W. S. Fisher, eds. 2000. Evaluation Guidelines for Ecological
4426 Indicators. EPA/620/R-99/005. U.S. Environmental Protection Agency, Office of
4427 Research and Development, Research Triangle Park, NC. p. 107.
4428
4429 Jaworski, E. and C. N. Raphael. 1979. Impact of Great Lakes Water Level Changes on Coastal
4430 Wetlands. Institute of Water Research, Michigan State University, East Lansing, MI.
4431
4432 Jaworski, N. A., R. W. Howarth, L. J. Hetling. 1997. Atmospheric deposition of nitrogen oxides
4433 onto the landscape contributes to coastal eutrophication in the Northeast United States.
4434 Environ. Sci. Techno. 31:1995-2004.
4435
4436 Jensen, J. E., S. R. Cooper, and C. J. Richardson. 1999. Calibration of modern pollen along a
4437 nutrient gradient in Everglades Water Conservation Area-2A. Wetlands 19:675-688.
4438
4439 Johansen, R. C., J. C. Imhoff, J. L. Kittle, Jr. and A. S. Donigian, 1984. Hydrological Simulation
4440 Program - FORTRAN (HSPF): Users Manual for Release 8.0, EPA-600/3-84-066,
4441 Environmental Research Laboratory, U.S. EPA, Athens, GA. 30613.
4442
4443 Johnston C. A., N. E Detenbeck, J. P. Bonde, and G. J. Niemi. 1988. Geographic information
4444 systems for cumulative impact assessment. Photogr Engin Remote Sens 54:1609-1615.
4445
4446 Johnston C. A., N. E. Detenbeck, and G. J. Niemi. 1990. The cumulative effect of wetlands on
4447 stream water quality and quantity: A landscape approach. Biogeochemistry 10:105-141.
4448
4449 Johnston C. A. 1991. Sediment and nutrient retention by freshwater wetlands: effects on surface
4450 water quality. Critical Rev Environ Control 21:491-565.
4451
4452 Jude, D. F., and J. Pappas. 1992. Fish utilization of Great Lakes coastal wetlands. Journal of
4453 Great Lakes Research 18: 651-672.
4454
4455 Kadlec, R. H. 1999. The limits of phosphorus removal in wetlands. Wetlands Ecology and
4456 Management 7: 165-175.
4457
4458 Kadlec, R. H. and F. B. Bevis. 1990. Wetlands and wastewater: Kinross, Michigan. Wetlands
4459 10:77-92.
4460
4461 Karr, J. R. and E. W. Chu. 1999. Restoring Life in Running Waters. Island Press, Washington,
4462 DC.
4463
123
-------
December 2006- DRAFT REFERENCES
4464 Kentucky Division of Water. 1993. Methods for Assessing Biological Integrity of Surface
4465 Waters. Kentucky Natural Resources and Environmental Protection Cabinet, Frankfort,
4466 KY.
4467
4468 Kentula, M. E., R. P. Brooks, S. E. Gwin, C. C. Holland, A. D. Sherman, and J. C. Sifneos. 1993.
4469 An Approach to Improving Decision Making in Wetland Restoration and Creation.
4470 Hairston, AJ.,ed. CK Smoley Inc. CRC Press, Inc., Boca Raton, FL.
4471
4472 Keough, J. R., M. Sierszen, and C. A Hagley. 1996. Analysis of a Lake Superior coastal food
4473 web, using stable isotope techniques. Limnol. Oceanogr. 41:136-146.
4474
4475 Keough, J. R., T. Thompson, G. R. Guntenspergen, and D. Wilcox. 1999. Hydrogeomorphic
4476 factors and ecosystem responses in coastal wetlands of the Great Lakes. Wetlands
4477 19:821-834.
4478
4479 Kerans, B. L. and J. R. Karr. 1994. A benthic index of biotic integrity (B-IBI) for rivers of the
4480 Tennessee Valley. Ecol. Appl. 4:768-785.
4481
4482 Keys, J. E., Jr., C. A. Carpenter, S. L. Hooks, F. Koenig, W. H. McNab, W. E. Russell, and M. L.
4483 Smith. 1995. Ecological Units of the Eastern United States: First Approximation. U.S.
4484 Department of Agriculture, Forest Service.
4485
4486 King, R. S. and C. J. Richardson. 2002. Evaluating subsampling approaches and
4487 macroinvertebrate taxonomic resolution for wetland bioassessment. Journal of the North
4488 American Benthological Society 21(1):150-171.
4489
4490 Kitting, C. L., B. Fry, and M. D. Morgan. 1984. Detection of inconspicuous epiphytic algae
4491 supporting food webs in seagrass meadows. Oecologia, 62:145-149.
4492
4493 Koerselman, W. and J. T. A. Verhoeven. 1995. Eutrophication of fen ecosystems: external and
4494 internal nutrient sources and restoration strategies. P.91-112. In: Restoration of
4495 Temperate Wetlands. Wheeler, B.D., S.C. Shaw, W.I. Fojt, and R.A. Robertson (eds.),
4496 John Wiley and Sons, Chichester, England.
4497
4498 Koerselman, W. and A. F. M. Meuleman. 1996. The vegetation N:P ratio: a new tool to detect
4499 the nature of nutrient limitation. Journal of Applied Ecology 33(6): 1441-1450.
4500
4501 Kuo, S. 1996. Phosphorus. In: Methods of Soil Analysis. Part 3, Chemical Methods B SSSA Book
4502 Series No. 5. D. L. Sparks et al. (eds.), Soil Science Society of America Inc., pp. 869-919.
4503
4504 Lane, C.R. 2000. Proposed wetland regions for Florida freshwater wetlands. 122 pp. available at:
4505 http://www.dep.state.fl.us/labs/library/index.htm
124
-------
December 2006- DRAFT REFERENCES
4506
4507 Latour, J. B. and R. Reiling. 1993. A multiple stress model for vegetation ('move'): A tool for
4508 scenario studies and standard-setting. Science of the Total Environment, 1993.
4509
4510 Latour, J. B., R. Reiling, and W. Slooff. 1994. Ecological standards for eutrophication and
4511 dessication: Perspectives for a risk assessment. Water, Air, and Soil Pollution 78(3-
4512 4):265-77.
4513
4514 Leibowitz SG, Abbruzzese A, Adamus PR, Hughes LE, Irish JT. 1992. A Synoptic Approach to
4515 Cumulative Impact Assessment: A Proposed Methodology. Environmental Research
4516 Laboratory, U.S. Environmental Protection Agency, Corvallis, OR. EPA/600/R-92/167.
4517
4518 Lohman, K., J. R. Jones, and C. Baysinger-Daniel. 1991. Experimental Evidence for Nitrogen
4519 limitation in an Ozark Stream. J. N. Am. Benthol. Soc. 10:13-24.
4520
4521 Lougheed, V. L., B. Crosbie, and P. Chow-Fraser. 2001. Primary determinants of macrophyte
4522 community structure in 62 marshes across the Great Lakes basin: latitude, land use, and
4523 water quality effects. Canadian Journal of Fisheries and Aquatic Sciences 58: 1603-1612.
4524
4525 Mann, K. H. 1972. Macrophyte production and detritus food chains in coastal waters. Mem. 1st.
4526 Ital. Idrobiol. Suppl. 29: 353-383.
4527
4528 Mann, K. H. 1988. Production and use of detritus in various freshwater, estuarine, and coastal
4529 marine ecosystems. Limnol. Oceanogr. 33: 910-930.
4530
4531 Martin, J. F, and K. R. Reddy. 1997. Interaction and spatial distribution of wetland nitrogen
4532 processes. Ecological Modelling 105: 1-21.
4533
4534 Maurer, D. I. and J. B. Zedler. 2002. Differential invasion of a wetland grass explained by tests
4535 of nutrients and light availability on establishment and clonal growth. Oecologia
4536 131:279-88.
4537
4538 Maxwell, J. R., C. J. Edwards, M. E. Jensen, S J. Paustian, H. Parott, and D. M. Hill. 1995. A
4539 hierarchical framework of aquatic ecological units in North America (Nearctic Zone).
4540 USD A, Forest Service, Technical Report NC-176.
4541
4542 McCormick, P. V., and J. Cairns, Jr. 1994. Algae as indicators of environmental change. J. Appl.
4543 Phycol. 6:509-526.
4544
4545 McCormick, P. V., P. S. Rawlik, K. Lurding, E. P. Smith, F. H. Sklar, 1996. Periphyton-water
4546 quality relationships along a nutrient gradient in the northern Florida Everglades. Journal
4547 of the North American Benthological Society 15: 433-449.
125
-------
December 2006- DRAFT REFERENCES
4548
4549 McCormick P. V., S. Newman, S. Miao, R. Reddy, D. Gawlick, C. Fitz, T. Fontaine, and D.
4550 Marley. 1999. Ecological Needs of the Everglades. In: Redfield G (ed). Everglades
4551 Interim Report. South Florida Water Management District, West Palm Beach, FL.
4552
4553 McDougal, R. L., L. G. Goldsborough, and B. J. Hann. 1997. Responses of a prairie wetland to
4554 press and pulse additions of inorganic nitrogen and phosphorus: production by planktonic
4555 and benthic algae. Archiv fuer Hydrobiologie [Arch. Hydrobiol.]. 140(2): 145-67.
4556
4557 McElroy, A. D., S. W. Chiu, J. W. Nabgen, A. Aleti, and R. W. Bennett. 1976. Loading
4558 functions for assessment of water pollution for non-point sources. EPA 600/2-76/151.
4559 U.S. Environmental Protection Agency, Washington, D.C.
4560
4561 McKee, P. M., T. R. Batterson, T. E. Dahl, V. Glooschenko, E. Jaworski, J. B. Pearce, C. N.
4562 Raphael, T. H. Whillans, and E. T. LaRoe. 1992. Great Lakes aquatic habitat
4563 classification based on wetland classification systems, Ch. 4, In: The Development of an
4564 Aquatic Habitat Classification System for Lakes. W. Dieter, N. Busch, and P. G. Sly
4565 (eds.) CRC Press, Ann Arbor, MI.
4566
4567 McJannet, C. L., P. A. Keddy, and F. R. Pick. 1995. Nitrogen and phosphorus tissue
4568 concentrations in 41 wetland plants: A comparison across habitats and functional groups.
4569 Functional Ecology 9:231-238.
4570
4571 Mendelssohn, I. A. and D. Burdick. 1988. The relationship of soil parameters and root
4572 metabolismto primary production in periodically inundated soils. In: The Ecology and
4573 Management of Wetlands. Volume 1: Ecology of Wetlands, Hook, D.D. et. al. (ed.),
4574 Timber Press, Portland Oregon, pp. 398-428.
4575
4576 Michigan Natural Features Inventory. 1997. Great Lakes coastal wetlands: An overview of
4577 controlling abiotic factors, regional distribution, and species composition. Michigan
4578 Natural Features Inventory, Lansing, MI. USEPA Grant #GL9 95810-02.
4579
4580 Mills, W.B. 1985. Water quality assessment: A screening procedure for toxic and conventional
4581 pollutants in surface and ground water. EPA/600/6-85/002a. Environmental Research
4582 Laboratory, U.S. Environmental Protection Agency, Athens, GA.
4583
4584 Miltner, R. J. and E. T. Rankin. 1998. Primary nutrients and the biotic integrity of rivers and
4585 streams. Freshwater Biol. 40:145-158.
4586
4587 Mitsch W. J., J. G. Gosselink 1993. Wetlands New York: Van Nostrand Reinhold.
4588
4589 Mitsch, W. J. and J. G. Gosselink. 2000. Wetlands, Van Nostrand Reinhold, New York.
126
-------
December 2006- DRAFT REFERENCES
4590
4591 Mitsch, W. J., and N. Wang. 2000. Large-scale coastal wetland restoration on the Laurentian
4592 Great Lakes: Determining the potential for water quality improvement. Ecological
4593 Engineering 15: 267-28.
4594
4595 Moore, L. W., C. Y. Chew, R. H. Smith, and S. Sahoo. 1992. Modeling of best management
4596 practices on North Reelfoot Creek, Tennessee. Water Environment Research 64:
4597 241-247.
4598
4599 Morris, J. T. and P. M. Bradley. 1999. Effects of nutrient loading on the carbon balance of
4600 coastal wetland sediments. Limnology and Oceanography 44(3): 699-702.
4601
4602 Mudroch, A. and J. A. Capobianco. 1979. Effects of treated effluent on a natural marsh. Journal
4603 of the Water Pollution Control Federation 51:2243-2256.
4604
4605 Mulholland, P. J., A. D. Steinman, A. V. Palumbo, J. W. Elwood, D. B. Kirschtel. 1991. Role of
4606 Nutrient Cycling and Herbivory in Regulating Periphyton Communities in Laboratory
4607 Stream. Ecology 72: 966-982.
4608
4609 Mulvaney, R. L. 1996. Nitrogen - Inorganic forms. In: Methods of Soil Analyis Part 3. Chemical
4610 Methods, J. M. Bigham et al. (eds). Soil Science Society of America, p. 1123.
4611
4612 Murkin, H. R.; J. B. Pollard, M. P. Stainton, J. A. Boughen, and R. D. Titman. 1994. Nutrient
4613 additions to wetlands in the Interlake region of Manitoba, Canada: Effects of periodic
4614 additions throughout the growing season. Hydrobiologia 279-280: 483-495.
4615
4616 Nair, V. D., D. A. Graetz, and K. M. Portier. 1995. Forms of phosphorus in soil profiles from -
4617 dairies of south Florida. Soil Sci. Soc. Am. J. 59:1244-1249.
4618
4619 Neill, C. 1990. Effects of nutrients and water levels on species composition in prairie whitetop
4620 (Scolochloafestucaced) marshes. Can. J. Bot. 68: 1015-1020.
4621
4622 Neill, C. 1990. Effects of nutrients and water levels on emergent macrophyte biomass in a prairie
4623 marsh. Can. J. Bot. 68: 1007-1014.
4624
4625 Neill, C. 1992. Life history and population dynamics of whitetop (Scolochloafestucaced) shoots
4626 under different levels of flooding and nitrogen supply. Aquatic Botany 42: 241-252.
4627
4628 Nelson, D. W., and L. E. Sommers, 1996. Total carbon, organic carbon, and organic matter. In:
4629 Methods of Soil Analysis. Part 3, Chemical Methods - SSSA Book Series No. 5. D. L.
4630 Sparks et al. (eds), Soil Science Society of America Inc., pp. 961-1010.
4631
127
-------
December 2006- DRAFT REFERENCES
4632 Neter, J., W. Wasserman, and M. Kutner. 1983. Applied Linear Regression Models., Richard D.
4633 Irwin Inc., Homewood, IL.
4634
4635 Newbould, P. J. 1967. Methods of estimating the primary production of forests. International
4636 Biological Programme handbook no. 2. Oxford and Edinburgh. Blackwell Scientific
4637 Publications.
4638
4639 Newell, R. I. E., N. Marshall, A. Sasekumar, and V. C. Chong. 1995. Relative importance of
4640 benthic microalgae, phytoplankton, and mangroves as sources of nutrition for penaeid
4641 prawns and other coastal invertebrates from Malaysia. Mar. Biol. 123: 595-606.
4642
4643 Nichols, D. S. 1983. Capacity of natural wetlands to remove nutrients from wastewater. Journal
4644 of the Water Pollution Control Federation 55:495-505.
4645
4646 Nicholson, B. J. 1995. The wetlands of Elk Island National Park: Vegetation classification, water
4647 chemistry, and hydrotopographic relationships. Wetlands 15: 119-133.
4648
4649 Nixon, S. 1980. Between coastal marshes and coastal waters - a review of twenty years of
4650 speculation and research on the role of salt marshes in estuarine productivity and water
4651 chemistry. Pages 437-525 in P. Hamilton and K. B. MacDonald, editors. Estuarine and
4652 wetland processes with emphasis on modeling. Plenum Press.
4653
4654 Noe, G.B., D. L. Childers, and R. D. Jones. 2001. Phosphorus biogeochemistry and the impact of
4655 phosphorus enrichment: Why is the Everglades so unique? Ecosystems 4:603-24.
4656
4657 NRCS, 1998. Indicators of Hydric Soils in the United States.
4658
4659 Odum, E. P., and de al A. A. Cruz. 1967. Particulate organic detritus in a Georgia salt marsh-
4660 estuarine ecosystem. In: Estuaries. G. H. Lauff, ed. Publ. Am. Assoc. Adv. Sci. 83: 383-
4661 388.
4662
4663 OECD. 1982. Eutrophication of Waters: Monitoring Assessment and Control. OECD, Paris.
4664 pp.154
4665
4666 Olde Venterink, H., and M. J. Wassen. 1997. A comparison of six models predicting vegetation
4667 response to hydrological habitat change. Ecological Modelling 101:347-361.
4668
4669 Olde Venterink, H., N. M. Pieterse, J. D. M. Belgers, M. J. Wassen and P. C. DeRuiter. 2002. N,
4670 P, and K budgets along nutrient availability and productivity gradients in wetlands.
4671 Ecological Applications 12:1010-1026.
4672
4673 Omernik, J.M. 1987. Ecoregions of the conterminous United States. Ann Assoc Am Geogr
4674 77:118-125.
128
-------
December 2006- DRAFT REFERENCES
4675
4676 Omernik, J. M., M. A. Shirazi, and R. M. Hughes. 1982. A synoptic approach for regionalizing
4677 aquatic ecosystems. In: In-place resource inventories: principles and practices,
4678 proceedings of a national workshop. August 9-14, 1981. Univ. of Maine, Orono, Maine.
4679 Society of American Foresters, pp. 199-218.
4680
4681 Palik A. J., C. P. Goebel, K. L. Kirkman, and L. West. 2000. Using landscape hierarchies to
4682 guide restoration of disturbed ecosystems. Ecol Appl 10(1): 189-202.
4683
4684 Pauli, D., M. Peintinger, and B. Schmid. 2002. Nutrient enrichment in calcareous fens: effects on
4685 plant species and community structure. Basic and Applied Ecology 3:255-66.
4686
4687 Paulsen, S. G., D. P. Larsen, P. R. Kaufmann, T. R. Whittier, J. R. Baker, D. V. Peck, J. McGue,
4688 R. M. Hughes, D. McMullen, D. Stevens, J. L. Stoddard, J. Larzorchak, W. Kinney, A. R.
4689 Selle, and RHjort. 1991. Environmental Monitoring and Assessment Program (EMAP)
4690 Surface Waters Monitoring and Research Strategy B Fiscal Year 1991. U.S.
4691 Environmental Protection Agency, Office of Research and Development, Washington,
4692 D.C. and Environmental Research Laboratory, Corvallis, Oregon. EPA-600-3-91-002.
4693
4694 Phillips, G. L., D. Eminson, and B. Moss. 1978. A mechanism to account for macrophyte decline
4695 in progressively eutrophicated freshwaters. Aquatic Botany 4:103-26.
4696
4697 Phillips, J. D. 1990. Saturation-based model of relative wetness for wetland identification. Water
4698 Resources Bulletin 26:333-342.
4699
4700 Pielou, E. C. 1984. The Interpretation of Ecological Data: A Primer on Classification and
4701 Ordination. Wiley, New York.
4702
4703 Poiani, K. A., and W. C. Johnson. 1993. A spatial simulation model of hydrology and vegetation
4704 dynamics in semi-permanent prairie wetlands. Ecological Applications [Ecol. Appl.],
4705 3(2): 279-293.
4706
4707 Poiani, K. A., W. C. Johnson, G. A. Swanson, and T. C. Winter. 1996. Climate change and
4708 northern prairie wetlands: Simulations of long-term dynamics. Limnol. Oceanogr. 41:
4709 871-881.
4710
4711 Preston, S. D. and J. W. Brakebill. 1999. Application of Spatially Referenced Regression
4712 Modeling for the Evaluation of Total Nitrogen Loading in the Chesapeake Bay
4713 Watershed (U.S. Geological Survey Water Resources Investigations Report 99-4054,
4714 Baltimore, Maryland).
4715
4716 Qian, S. S., R. S. King, and C. J. Richardson. 2003. Two statistical methods for detecting
4717 environmental thresholds. Ecological Modelling 166: 87-97.
4718
129
-------
December 2006- DRAFT REFERENCES
4719 Raschke, R. 1993. Guidelines for assessing and predicting eutrophication status of small
4720 southeastern piedmont impoundments. EPA Region IV. Environmental Services
4721 Division, Ecological Support Branch, Athens, GA
4722
4723 Reddy, K. R., and R. D. Delaune. 2007. Biogeochemistry of Wetlands: Science and Applications.
4724 CRC Press, Boca Raton, Fl. (in press).
4725
4726 Reddy K. R., R. D. DeLaune, W. F. DeBusk, and M. S. Koch. 1993. Long-term nutrient
4727 accumulation rates in the Everglades. Soil Sci Soc Am J 57:1147-1155.
4728
4729 Reddy, K. R., O. A. Diaz, L. J. Scinto, and M. Agami. 1995. Phosphorus dynamics in selected
4730 wetlands and streams of the Lake Okeechobee Basin. Ecol. Eng. 5:183-208.
4731
4732 Reddy, K. R., Y. Wang, W. F. DeBusk, M. M. Fisher and S. Newman. 1998. Forms of soil
4733 phosphorus in selected hydrologic units of Florida Everglades ecosystems. Soil Sci. Soc.
4734 Am. 1.62:1134-1147.
4735
4736 Reddy, K. R., J. R. White, A. Wright, and T. Chua. 1999. Influence of phosphorus loading on
4737 microbial processes in the soil and water column of wetlands. Chapter 10 In: Phosphorus
4738 Biogeochemistry of Subtropical Ecosystems: Florida as a Case Example. Reddy, K.R,
4739 GA. O'Connor, and C.L. Schelske (eds.) CRC/Lewis Pub. pp. 249-273.
4740
4741 Richardson, C. J. and S. S. Qian. 1999. Long-term phosphorus assimilative capacity in
4742 freshwater wetlands: A new paradigm for sustaining ecosystem structure and function.
4743 Environmental Science & Technology 33:1545-1551.
4744
4745 Richardson, C. J., S. Qian, C. B. Craft, and R. G. Quails. 1997. Predictive models for phosphorus
4746 retention in wetlands. Wetlands Ecology and Management 4:159-175.
4747
4748 Robertson, D. M., D. A. Saad, and A. M. Wieben. 2001. An Alternative Regionalization Scheme
4749 for Defining Nutrient Criteria for Rivers and Streams. USGS Water-Resources
4750 Investigations Report 01 -4073 .USGS, Middleton, WI.
4751
4752 Rose, C. and W. G. Crumpton, 1996. Effects of emergent macrophytes on dissolved oxygen
4753 dynamics in a prairie pothole wetland. Wetlands 16:495-502.
4754
4755 Rosgen, D. 1996. Applied River Morphology. Wildland Hydrology, Pagosa Springs, CO. 380 pp.
4756
4757 Rybczyk, J. M., G. Garson, and J. W. Jr. Day. 1996. Nutrient enrichment and decomposition in
4758 wetland ecosystems: models, analyses and effects. Current Topics Wetland Biogeochem.
4759 2:52-72.
4760
130
-------
December 2006- DRAFT REFERENCES
4761 Sandilands, K. A., B. J. Hann, and L. G. Goldsborough. 2000. The impact of nutrients and
4762 submersed macrophytes on invertebrates in a prairie wetland, Delta Marsh, Manitoba.
4763 Archiv fuer Hydrobiologie 148:441-59.
4764
4765 SFWMD, \999.EvergladesInterimReport, 1999. South Florida Water Management District,
4766 West Palm Beach, FL
4767
4768 SFWMD, 2000. Everglades Consolidated Report, 2000. South Florida Water Management
4769 District, West Palm Beach, FL
4770
4771 SFWMD, 2001. Everglades Consolidated Report, 200 J. South Florida Water Management
4772 District, West Palm Beach, FL
4773
4774 Shaver, G. R. and J. M. Melillo. 1984. Nutrient budgets of marsh plants: Efficiency concepts
4775 and relation to availability. Ecology 65:1491-1510.
4776
4777 Shaver G. R., L. C. Johnson, D. H. Cades, G. Murray, J. A. Laundre, E. B. Rastetter, K. J.
4778 Nadelhoffer, and A. E. Giblin. 1998. Biomass and CC>2 flux in wet sedge tundras:
4779 responses to nutrients, temperature and light. Ecol. Monogr. 68:75-97.
4780
4781 Shaw, S. P. and C. G. Fredine. 1956. Wetlands of the United States. U.S. Fish and Wildlife
4782 Service, Circ. 39.
4783
4784 Smith, R. D., A. Ammann, C. Bartoldus, and M. M. Brinson. 1995. An approach for assessing
4785 wetland functions using hydrogeomorphic classification, reference wetlands, and
4786 functional indices. U.S. Army Corps of Engineers, Waterways Experiment Station.
4787 Wetlands Research Program Technical Report WRP-DE-9.
4788
4789 Smith, R.A., G.E. Schwarz, and R.B. Alexander, 1997, Regional interpretation of water-quality
4790 monitoring data, Water Resources Research, v. 33, no. 12, pp. 2781-2798
4791
4792 Smith, R. A., R. B Alexander, and Gregory E. Schwarz. 2003. Natural background
4793 concentrations of nutrients in streams and rivers of the conterminous United States.
4794 Environmental Science and Technology v. 37, no. 14, pp. 3039-3047.
4795
4796 Spieles, D. J. and W. J. Mitsch. 2000. Macroinvertebrate community structure in high- and
4797 low-nutrient constructed wetlands. Wetlands 20:716-729.
4798
4799 Stephenson, M., G. Turner, P. Pope, J. Colt, A. Knight, and G. Tchobanoglous. 1980. Appendix
4800 A: The environmental requirements of aquatic plants. In: The Use and Potential of
4801 Aquatic Species for Wastewater Treatment, California State Water Resources Control
4802 Board, Sacramento, CA, USA.
131
-------
December 2006- DRAFT REFERENCES
4803
4804 Stevenson, R. J. 1984. Epilithic and epipelic diatoms in the Sandusky River, with emphasis on
4805 species diversity and water quality. Hydrobiologia 114:161-175.
4806
4807 Stevenson, R J. 1996. An introduction to algal ecology in freshwater benthic habitats. In: Algal
4808 Ecology: Freshwater Benthic Ecosystems. Stevenson, R. J., M. Bothwell, and R.L. Lowe
4809 (eds.). Academic Press, San Diego, CA. pp. 3-30.
4810
4811 Stevenson, R. J. 1997. Scale dependent determinants and consequences of benthic algal
4812 heterogeneity. J. N. Am. Benthol. Soc. 16(l):248-262.
4813
4814 Stevenson, R. J. 1998. Diatom indicators of stream and wetland stressors in a risk management
4815 framework. Environ. Monitor. Assess. 51:107-118.
4816
4817 Stevenson, R. J. and Y. Pan. 1999. Assessing ecological conditions in rivers and streams with
4818 diatoms. In: The Diatoms: Applications to the Environmental and Earth Sciences.
4819 Stoermer, E. F. and J. P. Smol (eds.). Cambridge University Press, Cambridge, UK. pp.
4820 11-40.
4821
4822 Stevenson, R. J. and F. R. Hauer. 2002. Integrating Hydrogeomorphic and Index of Biotic
4823 Integrity approaches for environmental assessment of wetlands. Journal of the North
4824 American Benthological Society 21:502-513.
4825
4826 Stevenson, R. J. and J. P. Smol. 2003. Use of algae in environmental assessment. In: Freshwater
4827 Algae in North America: Classification and Ecology. Wehr, J. D. and R. G. Sheath (eds.).
4828 pp. 775-804. Academic Press, San Diego, CA.
4829
4830 Stevenson, R. J., P. V. McCormick and R. Frydenborg. 2001. Methods for Evaluating Wetland
4831 Condition: Using Algae to Assess Environmental Conditions in Wetlands.
4832 EPA-843-B-00-002k. U.S. Environmental Protection Agency, Office of Water,
4833 Washington D. C.
4834
4835 Stevenson, R. J., Y. Pan, and P. Vaithiyanathan. 2002. Ecological assessment and indicator
4836 development in wetlands: the case of algae in the Everglades, USA. Verhandlungen
4837 Internationale Vereinigung fur Theoretische und Andgewandte Limnologie 28:1248-
4838 1252.
4839
4840 Stevenson, R. J., B. C. Bailey, M. C. Harass, C. P. Hawkins, J. Alba-Tercedor, C. Couch, S.
4841 Dyer, F. A. Fulk, J. M. Harrington, C. T. Hunsaker, and R. K. Johnson. 2004a. Designing
4842 data collection for ecological assessments. In: M. T. Barbour, S. B. Norton, H. R.
4843 Preston, and K. W. Thornton, eds. Ecological Assessment of Aquatic Resources: Linking
4844 Science to Decision-Making. Pgs 55-84. Society of Environmental Toxicology and
4845 Chemistry, Pensacola, Florida. ISBN 1-880611-56-2.
132
-------
December 2006- DRAFT REFERENCES
4846
4847 Stevenson, R. 1, B. C. Bailey, M. C. Harass, C. P. Hawkins, J. Alba-Tercedor, C. Couch, S.
4848 Dyer, F. A. Fulk, J. M. Harrington, C. T. Hunsaker, and R. K. Johnson. 2004b.
4849 Interpreting results of ecological assessments. In: M. T. Barbour, S. B. Norton, H. R.
4850 Preston, and K. W. Thornton, eds. Ecological Assessment of Aquatic Resources: Linking
4851 Science to Decision-Making. Pgs 85-111. Society of Environmental Toxicology and
4852 Chemistry, Pensacola, Florida. ISBN 1-880611-56-2.
4853
4854 Stewart, R. E. and H. A. Kantrud. 1971. Classification of natural ponds and lakes in the glaciated
4855 prairie region, U.S. Fish and wildlife Service Research Pub. 92.
4856
4857 Stewart-Oaten, A., W. W. Murdoch, and K.R Parker. 1986. Environmental impact assessment:
4858 "Pseudoreplication" in time? Ecology 67:929-940.
4859
4860 Stoddard, J. L., D. P. Larsen, C. P. Hawkins, R. K. Johnson, and R. H. Norris. 2006. Setting
4861 expectations for the ecological condition of streams: the concept of reference condition.
4862 Ecological Applications 16:1267-1276.
4863
4864 Sullivan, M. J., and C. A. Moncreiff. 1990. Edaphic algae are an important component of salt
4865 marsh food-webs: evidence from multiple stable isotope analyses. Mar. Ecol. Progr. Ser.
4866 62: 149-159.
4867
4868 Suter G. W. 1993. Ecological Risk Assessment. Boca Raton, FL: Lewis Publishers.
4869
4870 Svengsouk, L. J. and W. J. Mitsch. 2001. Dynamics of mixtures of Typha latifolia and
4871 Schoenoplectus tabernaemontani in nutrient enrichment wetland experiments. American
4872 Midland Naturalist 145:309-24.
4873
4874 Tate, C. M. 1990. Patterns and controls of nitrogen in tallgrass prairie streams. Ecology
4875 71:20007-2018.
4876
4877 Thomas, G. W. 1996. Soil pH and soil acidity. In: Methods of Soil Analysis. Part 3, Chemical
4878 Methods-SSSA Book Series No. 5. D.L. Sparks et al. (eds.), Soil Science Society of
4879 America Inc., pp. 475-490.
4880
4881 Thomas, C. R. and R. R. Christian. 2001. Comparison of nitrogen cycling in salt marsh zones
4882 related to sea level rise. Marine Ecology - Progress Series 221:1-16.
4883
4884 Thomas, L. and C. J. Krebs. 1997. A review of statistical power analysis software. Bulletin of the
4885 Ecological Society of America 78(2): 126-139.
4886
133
-------
December 2006- DRAFT REFERENCES
4887 Thormann, M. N. and S.E. Bayley. 1997. Response of aboveground net primary plant production
4888 to nitrogen and phosphorus fertilization in peatlands in southern boreal Alberta, Canada.
4889 Wetlands 17:502-512.
4890
4891 Toth L. A., D. A. Arrington, M. A. Brady, and D. A. Muszick. 1995. Conceptual evaluation of
4892 factors potentially affecting restoration of habitat structure within the channelized
4893 Kissimmee River ecosystem. Restoration Ecol 3:160-180.
4894
4895 Trebitz, A. S., J. A. Morrice, A. M. Cotter. 2002. Relative Role of Lake and Tributary in
4896 Hydrology of Lake Superior Coastal Wetlands. Journal of Great Lakes Research 28:
4897 212-227.
4898
4899 Trexler, J. C. 1995. Restoration of the Kissimmee River—A conceptual model of past and
4900 present fish communities and its consequences for evaluating restoration success.
4901 Restoration Ecol 3:195-210.
4902
4903 Turner, R.E. and N. N. Rabalais. 1991. Changes in Mississippi River water quality this century
4904 and implications for coastal food webs. BioScience 41(3): 140-147.
4905
4906 Underwood, A. J. 1991. Beyond BACI: Experimental designs for detecting human
4907 environmental impacts on temporal variations in natural populations. Austral J Marine
4908 Freshw Res 42:569-587.
4909
4910 UrquhartN. S., S. G. Paulsen, and D. P. Larsen. 1998. Monitoring for policy-relevant regional
4911 trends over time. Ecol Appl 8(2):246-257.
4912
4913 USDA SCS. 1981. Land resource regions and major land resource areas of the United States.
4914 Agricultural Handbook 296. U.S. Government Printing Office. Washington, D.C. Map
4915 (scale 1:7,500,000) 156 pp.
4916
4917 USDA. 1999. Soil Taxonomy: A Basic System of Soil Classification for Making and Interpreting
4918 Soil Surveys. Agriculture handbook number 436. U.S. Department of Agriculture, Natural
4919 Resource Conseration Service. Washington, DC.
4920
4921 USEPA. 1971. Algal Assay Procedure: Bottle Test. National Eutrophi cation Research Program,
4922 U.S. Environmental Protection Agency Corvallis, OR.
4923
4924 USEPA. 1986. Quality Criteria for Water - 1986. Office of Water, U.S. Environmental
4925 Protection Agency, Washington DC. EPA440/5-86-001.
4926
4927 USEPA. 1990a. Agency Operating Guidance, FY 1991: Office of Water. Office of the
4928 Administrator, Washington, DC.
4929
134
-------
December 2006- DRAFT REFERENCES
4930 USEPA 1990b. Biological Criteria: National Program Guidance for Surface Waters. USEPA
4931 Office of Water. Washington, DC. EPA-440/5-90-004.
4932
4933 USEPA. 1993 a. Methods for chemical analysis of water and wastes. U.S. Environmental
4934 Protection Agency, Cincinnati, OH.
4935
4936 USEPA. 1993b. Guidance Specifying Management Measures for Sources of Nonpoint Pollution
4937 in Coastal Waters. Office of Water, U.S. Environmental Protection Agency, EPA 840-B-
4938 92-002.
4939
4940 USEPA. 1994. Water Quality Standards Handbook. SecondEd. Office of Water, U.S.
4941 Environmental Protection Agency. EPA 823-B-95-005.
4942
4943 USEPA. 1996. Biological Criteria: Technical Guidance for Streams and Small Rivers. Office of
4944 Water, Washington, DC. EPA 822-B-96-001.
4945
4946 USEPA. 1998a. National Strategy for the Development of Regional Nutrient Criteria,. Office of
4947 Water, U.S. Environmental Protection Agency. EPA 822-R-98-002.
4948
4949 USEPA. 1998b. National Water Quality Inventory: 1998 Report to Congress. Office of Water,
4950 U.S. Environmental Protection Agency. EPA 841-R-00-001.
4951
4952 USEPA. 1998c. Guidance for Quality Assurance Project Plans. Office of Research and
4953 Development, U.S. Environmental Protection Agency. EPA/600/R-98/018.
4954
4955 USEPA. 1999. Protocol for Developing Nutrient TMDLs. First Ed. November 1999. Watershed
4956 Branch of Assessment and Watershed Protection Division, Office of Water (4503F), U.S.
4957 Environmental Protection Agency. EPA 841-B-99-007.
4958
4959 USEPA. 2000a. Nutrient Criteria Technical Guidance Manual: Lakes and Reservoirs., First
4960 Edition. Office of Water. EPA-822-B-00-001.
4961
4962 USEPA. 2000b. Nutrient Criteria Technical Guidance Manual: Rivers and Streams. Office of
4963 Water, Office of Science and Technology. EPA-822-B-00-002.
4964
4965 USEPA. 2000c. National Water Quality Inventory 2000 Report. Office of Water, U. S.
4966 Environmental Protection Agency. EPA 841 -R-02-001.
4967
4968 USEPA. 2001. Nutrient Criteria Technical Guidance Manual: Estuarine and Coastal Marine
4969 Waters. Office of Water, Office of Science and Technology. EPA-822-B-01-003.
4970
135
-------
December 2006- DRAFT REFERENCES
4971 Valiela I. and J. M. Teal. 1974. Nutrient limitation in salt marsh vegetation. In: Ecology of
4972 Halophytes. R. J. Reimold, W. H. Queen, (eds). New York: Academic Press, pp.
4973 547-563.
4974
4975 Valiela, I, J. M. Teal and N. Y. Persson. 1976. Production and dynamics of experimentally
4976 enriched salt marsh vegetation: Belowground biomass. Limnology and Oceanography 29:
4977 245-252.
4978
4979 Van der Peijl, M. J., M. M. P van Oorschot, and J. T. A Verhoeven. 2000. Simulation of the
4980 effects of nutrient enrichment on nutrient and carbon dynamics in a river marginal
4981 wetland. Ecol. Model. 134: 169-184.
4982
4983 van der Valk, A. G. 2000. Vegetation dynamics and models. In Prairie Wetland Ecology: the
4984 Contributions of the Marsh Ecology Research Program. Murkin, H. R., A. G. van der
4985 Valk and W. R. Clark (Eds.), Iowa State University Press, Ames, IA. pp. 125-161.
4986
4987 Van Groenendael, J. M., M. J. M. van Mansfeld, A. J. M. Roozen, and V. Westhoff. 1993.
4988 Vegetation succession in lakes in the coastal fringe of West Connemara, Ireland. Aquatic
4989 Conservation: Marine and Freshwater Ecosystems 3: 25-41.
4990
4991 Verhoeven, J. T. A., and M. B. Schmitz. 1991. Control of plant growth by nitrogen and
4992 phosphorus in mesotrophic fens. Biogeochemistry 12:135-148.
4993
4994 Verhoeven, J. T. A., W. Koerselman, and B. Beltman. 1988. The vegetation of fens in relation to
4995 their hydrology and nutrient dynamics: a case study. In: Vegetation of Inland Waters.
4996 Handbook of Vegetation Science, J. J. Symoens, (ed.). Kluwer Academic Publishing,
4997 Dordrecht, Netherlands. 15:249-282
4998
4999 Vermeer, J. G. 1986. The effect of nutrient addition and lower the water table on the shoot
5000 biomass and species composition of a wet grassland community. Oecologia Plantarum
5001 7:145-155.
5002
5003 Vitousek, P. M., J. D. Aber, R. W. Howarth, G. E. Linkens, P. A. Matson, S. W. Schindler, W.
5004 H. Schlesinger and D. G. Tilman. 1997. Human alteration of the global nitrogen cycle:
5005 Sources and consequences. Ecological Applications 7(3): 737-750.
5006
5007 Voss, E. G. 1972. Michigan Flora. Part I. Gymnosperms andMonocots. Cranbrook Institute of
5008 Science Bulletin 55 and University of Michigan Herbarium. Bloomfield Hills, Michigan.
5009
5010 Wang, N., W. J. Mitsch. 2000. A detailed ecosystem model of phosphorus dynamics in
5011 created riparian wetlands, Ecological Modelling [Ecol. Model.]. 126 (2-3): 101-130.
5012
136
-------
December 2006- DRAFT REFERENCES
5013 Watershed Management Institute, Inc. 1998. Operation, maintenance, and management of
5014 stormwater management systems. Technical report with USEPA.
5015
5016 Weakley, A. S. and M. P. Schafale. 1991. Classification of pocosins of the Carolina coastal
5017 plain. Wetlands 11:355-375.
5018
5019 Wentz, W. A. 1976. The effects of sewage effluent on the growth and productivity of peatland
5020 plants. Ph.D. dissertation. The University of Michigan, Ann Arbor, MI, USA.
5021
5022 Wetzel, R.G. 2001. Fundamental processes within natural and constructed wetland ecosystems:
5023 short-term versus long-term objectives. Water Sci & Tech 44: 1-8.
5024
5025 Wharton, C. H., W. M. Kitchens, E. C. Pendleton. and T. W. Sipe. 1982. The Ecology of
5026 Bottomland Hardwood Swamps of the Southeast: A Community Profile, FWS/OBS-
5027 81/37, U.S. Fish and Wildlife Service, Biological Services Program, Washington, DC.
5028
5029 Whipple, S. A., Fleeger, J. W. andL. L. Cook. 1981. The Influence of Tidal Flushing, Light
5030 Exposure and Natant Macrofauna on Edaphic Chlorophyll a in a Louisiana Salt Marsh.
5031 Estuarine, Coastal and Shelf Science 13(6):637-643.
5032
5033 White, J. R and K. R. Reddy. 1999. The influence of nitrate and phosphorus loading on
5034 denitrifying enzyme activity in everglades wetland soils. Soil Sci. Soc. Am. J.
5035 63:1945-1954.
5036
5037 Whittaker, R. H. 1952. A study of summer foliage insect communities in the Great Smoky
5038 Mountains. Ecol. Monogr. 22:1-44.
5039
5040 Wilkinson, L. 1999. Systat 9.0 Statistics I. SPSS Inc. Chicago, IL.
5041
5042 Winter, T. C. 1977. Classification of the hydrogeologic settings of lakes in the north-central
5043 United States. Water Resources Research 13:753-767.
5044
5045 Winter, T. C. 1992. A physiographic and climatic framework for hydrologic studies of wetlands.
5046 In: Aquatic ecosystems in semi-arid regions: Implications for resource management.
5047 R.D. Roberts and M.L. Bothwell (eds.), NHRI Symp. Ser. 7. Saskatoon, Canada:
5048 Environment Canada, pp. 127-148.
5049
5050 Woo, I. and J. B. Zedler. 2002. Can nutrients alone shift a sedge meadow towards dominance by
5051 the invasive Typha x glaucal Wetlands 22:509-21.
5052
137
-------
December 2006- DRAFT REFERENCES
5053 Young, R., C. A. Onstad, D. D. Bosch, and W. P. Anderson. 1987. AGNPS: Agricultural
5054 Non-Point Source Pollution Model: A watershed analysis tool. USD A-Agricultural
5055 Research Service. Conservation Research Report 35, 77 pp.
5056
5057 Zheng, L.; Stevenson, R. J. and C. Craft. 2004. Changes In Benthic Algal Attributes During
5058 Salt Marsh Restoration. Wetlands 24: 309-323.
5059
5060 Zrum, L. and B. J. Hann. 2002. Invertebrates associated with submersed macrophytes in a prairie
5061 wetland: Effects of organophosphorus insecticide and inorganic nutrients. Archiv fuer
5062 Hydrobiologie 154:413-45.
138
-------
December 2006- DRAFT APPENDIX A: ACRONYM LIST AND GLOSSARY
5063 APPENDIX A. ACRONYM LIST AND GLOSSARY
5064
5065
5066 ACRONYMS
5067
5068 ACOE/ACE/COE - Army Corps of Engineers
5069 AGNPS - Agricultural Nonpoint Source Pollution model
5070 ARS - Agricultural Research Service
5071 BACI - Before/After, Control/Impact
5072 BMP - Best Management Practice
5073 BuRec - Bureau of Reclamation
5074 CCC - Commodity Credit Corporation
5075 CENR - Committee for the Environment and Natural Resources
5076 CGP - Construction General Permit
5077 CHN - Carbon-Hydrogen-Nitrogen
5078 CPGL - Conservation of Private Grazing Land
5079 CPP - Continuing Planning Process
5080 CREP - Conservation Reserve Enhancement Program
5081 CRP - Conservation Reserve Program
5082 CSO - Combined Sewer Overflow
5083 CWA - Clean Water Act
5084 CZARA - Coastal Zone Act Reauthorization Amendment
5085 DIP - Dissolved inorganic phosphorus
5086 DO - Dissolved oxygen
5087 DOP - Dissolved organic phosphorus
5088 DRP - Dissolved reactive phosphorus
5089 ECARP - Environmental Conservation Acreage Reserve Program
5090 ED AS - Ecological Data Application System
5091 Eh - Redox potential
5092 EMAP - Environmental Monitoring and Assessment Program
5093 EQIP - Environmental Quality Incentive Program
5094 FDEP - Florida Department of Environmental Protection
5095 FIP - Forestry Incentive Program
5096 GIS - Geographic Information System
5097 GPS - Geospatial Positioning System
5098 GWLF - Generalized Watershed Loading Function
5099 HEL - Highly credible land
5100 HGM - Hydrogeomorphic approach
139
-------
December 2006- DRAFT APPENDIX A: ACRONYM LIST AND GLOSSARY
5101 HSPF - Hydrologic Simulation Program - Fortran
5102 MFC A - Minnesota Pollution Control Agency
5103 NAAQS - National Ambient Air Quality Standard
5104 NASQAN - National Stream Quality Assessment Network
5105 NAWQA - National Water Quality Assessment
5106 NIS - Network Information System
5107 NIST - National Institute of Standards and Technology
5108 NOAA - National Oceanic and Atmospheric Administration
5109 NPDES - National Pollution Discharge Elimination System
5110 NPP - Net primary production
5111 NRCS - Natural Resources Conservation Service
5112 NSF - National Science Foundation
5113 NWI - National Wetlands Inventory
5114 OH EPA - Ohio EPA
5115 ONRW - Outstanding Natural Resource Waters
5116 PCB - Polychlorinated biphenyls
5117 PCS - Permit Compliance System
5118 PIP - Paniculate inorganic phosphorus
5119 POP - Paniculate organic phosphorus
5120 PSA - Particle size analysis
5121 QA/QC - Quality Assurance/Quality Control
5122 QC - Quality Control
5123 REMAP - Regional Environmental Monitoring and Assessment Program
5124 RF3-Reach File 3
5125 SCS - Soil Conservation Service
5126 SPARROW - Spatially Referenced Regressions on Watersheds
5127 SRP - Soluble reactive phosphorus
5128 STORET - Storage and Retrieval System
5129 SWAT - Soil and Water Assessment Tool
5130 TKN - Total Kjeldahl Nitrogen
5131 TMDL - Total Maximum Daily Load
5132 TP - Total Phosphorus
5133 TWINSPAN-
5134 USDA - United States Department of Agriculture
5135 USEPA - United States Environmental Protection Agency
5136 USFWS - United States Fish and Wildlife Service
5137 USGS - United States Geological Survey
5138 WEBB - Water, Energy, and Biogeochemical Budgets
5139 WHIP - Wildlife Habitat Incentive Program
5140 WLA - Wasteload Allocation
5141 WQBEL - Water Quality Based Effluent Limit
5142 WQS - Water Quality Standard
5143 WRP - Wetlands Reserve Program
140
-------
December 2006- DRAFT APPENDIX A: ACRONYM LIST AND GLOSSARY
5144 GLOSSARY
5145
5146 biocriteria
5147 (biological criteria) Narrative or numeric expressions that describe the desired biological condition of aquatic
5148 communities inhabiting particular types of waterbodies and serve as an index of aquatic community health. (USEPA
5149 1994).
5150
5151 cluster analysis
5152 An exploratory multivariate statistical technique that groups similar entities in an hierarchical structure.
5153
5154 criteria
5155 Elements of State water quality standards, expressed as constituent concentrations, levels, or narrative statements,
5156 representing a quality of water that supports a particular use. When criteria are met, water quality will generally
5157 protect the designated use (40 C.F.R. 131.3(b)).
5158
5159 designated use(s)
5160 Uses defined in water quality standards for each waterbody or segment whether or not the use is being attained
5161 (USEPA 1994).
5162
5163 detritus
5164 Unconsolidated sediments comprised of both inorganic and dead and decaying paniculate organic matter inhabited
5165 by decomposer microorganisms (Wetzel 1983).
5166
5167 ecological unit
5168 Mapped units that are delineated based on similarity in climate, landform, geomorphology, geology, soils,
5169 hydrology, potential vegetation, and water.
5170
5171 ecoregion
5172 A region defined by similarity of climate, landform, soil, potential natural vegetation, hydrology, and other
5173 ecologically relevant variables.
5174
5175 emergent vegetation
5176 "Erect, rooted herbaceous angiosperms that may be temporarily to permanently flooded at the base but do not
5177 tolerate prolonged inundation of the entire plant; e.g., bulrushes (Scirpus spp.), saltmarsh cordgrass" (Cowardin et
5178 al. 1979).
5179
5180 eutrophic
5181 Abundant in nutrients and having high rates of productivity frequently resulting in oxygen depletion below the
5182 surface layer (Wetzel 1983).
5183
5184 eutrophication
5185 The increase of nutrients in [waterbodies] either naturally or artificially by pollution (Goldman and Home 1983).
5186
5187 GIS (Geographical Information Systems)
5188 A computerized information system that can input, store, manipulate, analyze, and display geographically
5189 referenced data to support decision-making processes. (NDWP Water Words Dictionary)
5190
5191 HGM, hydrogeomorphic
5192 Land form characterized by a specific origin, geomorphic setting, water source, and hydrodynamic (NDWP Water
5193 Words Dictionary)
5194
141
-------
December 2006- DRAFT APPENDIX A: ACRONYM LIST AND GLOSSARY
5 195 index of biotic integrity (IBI)
5 196 An integrative expression of the biological condition that is composed of multiple metrics. Similar to economic
5 197 indexes used for expressing the condition of the economy.
5198
5199 interfluve
5200 An area of relatively unchannelized upland between adjacent streams flowing in approximately the same direction.
5201
5202 lacustrine
5203 "Includes wetlands and deepwater habitats with all of the following characteristics: (1)
5204 situated in a topographic depression or a dammed river channel; (2) lacking trees, persistent emergents, emergent
5205 mosses or lichens with greater than 30% areal coverage; and (3) total area exceeds 8 ha (20 acres). Similar wetland
5206 and deepwater habitats totaling less than 8 ha are also included in the Lacustrine System if an active wave-formed
5207 or bedrock shoreline feature makes up all or part of the boundary, or if the water depth in the deepest part of the
5208 basin exceeds 2 m (6.6 feet) at low water.. .may be tidal or nontidal, but ocean-derived salinity is always less than
5209 0.5%" (Cowardin et al. 1979).
5210
5211 lentic
5212 Relatively still-water environment (Goldman and Home 1983).
5213
5214 limnetic
5215 The open water of a body of fresh water.
5216
5217 littoral
5218 Region along the shore of a non-flowing body of water.
5219
5220 lotic
5221 Running-water environment (Goldman and Home 1983).
5222
5223 macrophyte (also known as SAV-Submerged Aquatic Vegetation)
5224 Larger aquatic plants, as distinct from the microscopic plants, including aquatic mosses, liverworts, angiosperms,
5225 ferns, and larger algae as well as vascular plants; no precise taxonomic meaning (Goldman and Home 1983).
5226
5227
5228 micrograms per liter, 10"6 grams per liter
5229
5230 mg/L
523 1 milligrams per liter, 10"3 grams per liter
5232
5233 mineral soil flats
5234 Level wetland landform with predominantly mineral soils
5235
5236 minerotrophic
5237 Receiving water inputs from groundwater, and thus higher in salt content (major ions) and pH than ombrotrophic
5238 systems.
5239
5240 mixohaline
5241 Water with salinity of 0.5 to 30%, due to ocean salts.
5242
5243 M
5244 Molarity, moles of an element as concentration
5245
142
-------
December 2006- DRAFT APPENDIX A: ACRONYM LIST AND GLOSSARY
5246 multivariate
5247 Type of statistics that relates one or more independent (explanatory) variables with multiple dependent (response)
5248 variables.
5249
5250 nutrient ecoregion
5251 Level II ecoregions defined by Omernik according to expected similarity in attributes
5252 affecting nutrient supply (http://www.epa.gov/OST/standards/ecomap.html)
5253
5254 oligotrophic
5255 Trophic status of a waterbody characterized by a small supply of nutrients (low nutrient release from sediments),
5256 low production of organic matter, low rates of decomposition, oxidizing hypolimnetic condition (high DO) (Wetzel
5257 1983).
5258
5259 palustrine
5260 "Nontidal wetlands dominated by trees, shrubs, persistent emergents, emergent mosses orlichens, and all such
5261 wetlands that occur in tidal areas where salinity due to ocean-derived salts is below 0.5%. It also includes wetlands
5262 lacking such vegetation, but with all of the following four characteristics: (1) area less than 8 ha (20 acres); (2)
5263 active wave-formed or bedrock shoreline features lacking; (3) water depth in the deepest part of basin less than 2 m
5264 at low water; and (4) salinity due to ocean-derived salts less than 0.5%" (Cowardin et al. 1979).
5265
5266 peatland
5267 "A type of wetland in which organic matter is produced faster than it is decomposed,
5268 resulting in the accumulation of partially decomposed vegetative material called Peat. In some mires peat never
5269 accumulates to the point where plants lose contact with water moving through mineral soil. Such mires, dominated
5270 by grasslike sedges, are called Fens. In other mires peat becomes so thick that the surface vegetation is insulated
5271 from mineral soil. These plants depend on precipitation for both water and nutrients. Such mires, dominated by acid
5272 forming sphagnum moss, are called Bogs." (NDWP Water Words Dictionary)
5273
5274 periphyton
5275 Associated aquatic organisms attached or clinging to stems and leaves of rooted plants or other surfaces projecting
5276 above the bottom of a waterbody (USEPA 1994).
5277
5278 pocosin
5279 Evergreen shrub bog, found on Atlantic coastal plain.
5280
5281 riverine wetland
5282 A hydrogeomorphic class of wetlands found in floodplains and riparian zones
5283 associated with stream or river channels.
5284
5285 slope wetland
5286 A wetland typically formed at a break in slope where groundwater discharges to
5287 the surface. Typically there is no standing water.
5288
5289 trophic status
5290 Degree of nutrient enrichment of a waterbody.
5291
5292 waters of the US
5293 Waters of the United States include:
5294 a. All waters that are currently used, were used in the past, or may be susceptible to use in interstate or foreign
5295 commerce, including all waters that are subject to the ebb and flow of the tide;
5296 b. All interstate waters, including interstate wetlands;
143
-------
December 2006- DRAFT APPENDIX A: ACRONYM LIST AND GLOSSARY
5297 c. All other waters such as interstate lakes, rivers, streams (including intermittent streams), mudflats, sandflats,
5298 wetlands, sloughs, prairie potholes, wet meadows, playa lakes, or natural ponds the use, degradation, or destruction
5299 of which would affect or could affect interstate or foreign commerce including any such waters:
5300 1 That are or could be used by interstate or foreign travelers for recreational or other
5301 purposes;
5302 2 From which fish or shellfish are or could be taken and sold in interstate or foreign
5303 commerce; or
5304 3 That are used or could be used for industrial purposes by industries in interstate
5305 commerce;
5306 d. All impoundments of waters otherwise defined as waters of the United States under this definition;
5307 e. Tributaries of waters identified in paragraphs (a) through (d) of this definition;
5308 f. The territorial sea; and
5309 g. Wetlands adjacent to waters (other than waters that are themselves wetlands) identified in paragraphs (a) through
5310 (f) of this definition.
5311
5312 wetland(s)
5313 Those areas that are inundated or saturated by surface or groundwater at a frequency and duration sufficient to
5314 support, and that under normal circumstances do support, a prevalence of vegetation typically adapted for life in
5315 saturated soil conditions [EPA, 40 C.F.R.§ 230.3 (t) / USACE,33 C.F.R. § 328.3 (b)].
5316
144
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
5317 APPENDIX B. CASE STUDY 1: DERIVING A PHOSPHORUS CRITERION FOR THE
5318 FLORIDA EVERGLADES
5319
5320 INTRODUCTION
5321
5322 The Everglades (Fig. 1) is the largest subtropical wetland in North America and is widely
5323 recognized for its unique ecological character. It has been affected for more than a century by
5324 rapid population growth in south Florida. Roughly half of the ecosystem has been drained and
5325 converted to agricultural and urban uses. Among other changes, the conversion of 500,000 acres
5326 of the northern Everglades to agriculture (the Everglades Agricultural Area or EAA) and the
5327 subsequent diking of the southern rim of Lake Okeechobee eliminated the normal seasonal flow
5328 of water southward from Lake Okeechobee, furthermore, the construction of a complex network
5329 of internal canals and levees disrupted the natural sheetflow of water through the system and
5330 created a series of impounded wetlands known as "Water Conservation Areas" or WCAs. This
5331 conversion from a hydrologically open to a highly managed wetland occurred gradually,
5332 beginning with the excavation of four major canals during the 1900-1910 period and culminating
5333 with the construction of the Central and South Florida Flood Control Project (CSFFCP) during
5334 the 1950s and 60s (Light and Dineen 1994).
5335
5336 The remnant Everglades is managed for multiple and often conflicting uses including water
5337 supply, flood control, and the hydrologic needs of the natural ecosystem. Water management
5338 operations have altered the quantity, quality, timing, and delivery of flows to the Everglades
5339 relative to the pre-disturbance system; some parts of the system have been damaged by
5340 overdrainage, excessive flooding in other areas has stressed native vegetation communities.
5341 Changes to the seasonal pattern of flooding and drying have influenced many ecological
5342 processes, including changes in the dominant micro- and macro-phytic vegetation, declines in
5343 critical species, and the nesting success of wading bird populations that rely on drawdowns
5344 during a narrow window of time to concentrate fish prey. Canal inputs containing runoff from
5345 agricultural and urban lands contribute roughly 50% of flows to the managed system and have
5346 increased loads of nutrients and contaminants. In particular, phosphorus (P) has been identified
5347 as a key limiting nutrient in the Everglades, and increased inputs of this nutrient have been
5348 identified as a significant factor affecting ecological processes and communities.
5349
5350 The primary source of P to the pre-disturbance Everglades was rainfall although seasonal flows
5351 from Lake Okeechobee likely contributed significant P to the northern fringe of the wetland.
5352 Prior to the implementation of P control efforts in the late 1990s, canal flows were estimated to
5353 contribute more than half of the P load to the managed Everglades (SFWMD 1992). Discharge
5354 from the EAA is the main source of water to the Everglades, with approximately 500,000 acre of
5355 farmland draining southward via SFWMD canals, and is the major source of anthropogenic P.
5356 Significant inputs also come from Lake Okeechobee, a naturally mesotrophic lake that has also
5357 been enriched by agricultural runoff. Several other agricultural and urban catchments contribute
5358
145
-------
December 2006- DRAFT
APPENDIX B: CASE STUDIES
5359
5360
5361
5362
5363
5364
5365
5366
5367
Appendix Bl. Figure Bl.l. Major hydrologic units of the remnant Florida Everglades (shaded
region) including (from north to south) the A.R.M. Loxahacthee National Wildlife Refuge
(LNWR), Water Conservation Area (WCA) 2A, WCA 3A, and Everglades National Park.
Shaded lines represent the regional canal and levee system that conveys water southward from
Lake Okeechobee and the Everglades Agricultural Area to the Everglades and urban areas along
the coast.
146
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
5368 smaller amounts of P via canal discharges into various parts of the Everglades. However, in
5369 general, canal P concentrations and loads (and associated wetland concentrations) decline from
5370 north to south .
5371
5372 The history of P enrichment and associated ecological impacts is not well documented, but
5373 probably occurred at a limited scale for much of the last century. Early reports by the South
5374 Florida Water Management District (e.g., Gleason et al. 1975, Swift and Nicholas 1987) showed
5375 an expansion of cattail and changes in the periphyton community in portions of the northern
5376 Everglades receiving EAA runoff. The severity and extent of P impacts were more fully
5377 recognized by 1988 when the Federal Government sued the State of Florida for allowing P-
5378 enriched discharges and associated impacts to occur in the Everglades. Settlement of this
5379 lawsuit eventually resulted in the enactment of the Everglades Forever Act by the Florida
5380 Legislature in 1994, which required the Florida Department of Environmental Protection (FDEP)
5381 to derive a numeric water quality criterion for P that would "prevent ecological imbalances in
5382 natural populations of flora or fauna" in the Everglades. These legal and legislative events
5383 provided the basis for numerous research and monitoring efforts designed to better understand
5384 the effects of P enrichment and to determine levels of enrichment that produced undesirable
5385 ecosystem changes.
5386
5387 Research and monitoring were initiated by the State of Florida (the Florida Department of
5388 Environmental Protection and the South Florida Water Management District) and other
5389 university research groups (e.g., Duke University, Florida International University, University of
5390 Florida) to better understand ecological responses to anthropogenic P inputs and to identify a P
5391 concentration or range of concentrations that result in unacceptable degradation of the
5392 Everglades ecosystem. This case study reviews research and monitoring conducted by the State
5393 to derive a P criterion for the Everglades. This criterion was proposed by the FDEP in 2001 and
5394 approved in 2003. This process is divided into 3 parts:
5395
5396 1. Defining the reference (i.e., historical) conditions for P and the oligotrophic ecology of
5397 the Everglades;
5398
5399 2. Determine the types of ecological impacts caused by P enrichment;
5400
5401 3. Identify wetland P concentrations that produce these impacts, and determine a
5402 criterion that will protect the resource from those impacts.
5403
5404
5405 DEFINING THE REFERENCE CONDITION
5406
5407 Several sources of information were used to characterize reference conditions across the
5408 Everglades. Sampling in minimally impacted locations (i.e., reference sites) believed to best
5409 reflect historical conditions provided the quantitative basis for establishing reference conditions
147
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
5410 with respect to P concentrations and associated ecological conditions. Where possible, this
5411 characterization was augmented by historical evidence. Written accounts of surveys conducted
5412 during the 1800s and early 1900s provided useful qualitative data on past ecological conditions.
5413 Early scientific literature contained substantial information on large-scale vegetation patterns
5414 (e.g., Davis 1943, Loveless 1959). Paleoecological assessments, including the dating and
5415 analysis of soil cores with respect to nutrient content and preserved materials such as pollen
5416 provided further information (e.g., Cooper and Goman 2001, Willard et al. 2001).
5417
5418 Predisturbance Everglades exhibited significant spatial and temporal variation, and, while its
5419 conversion to a smaller, more managed wetland resulted in the loss of some of this
5420 heterogeneity, the legacy of past variations in hydrology, chemistry, and biology remain in many
5421 areas. Legislation mandating the development of a P criterion stipulated that natural variation in
5422 P concentrations and ecological conditions within the remnant ecosystem be considered. This
5423 required that sampling efforts encompass the expected range of background variability in the
5424 remnant ecosystem. To ensure that spatial variation in P conditions were considered, sampling
5425 was conducted in all 4 major hydrologic units: The Loxahatchee National Wildlife Refuge
5426 (LNWR), WCA-2A, WCA-3A, and Everglades National Park (see Fig. 1).
5427
5428 Water Column Phosphorus
5429
5430 Nutrient inputs to the Everglades were historically derived primarily from atmospheric
5431 deposition (rainfall and dry fallout), which is typically low in P. Historical loading rates have
5432 been estimated from annual atmospheric P inputs in south Florida and reconstructions of P
5433 accumulation in Everglades soils and probably averaged less than 0.1 g P m^y"1 (SFWMD
5434 1992). Atmospheric inputs of P were augmented by inflows from Lake Okeechobee, which was
5435 connected by surface-water flows to the northern Everglades during periods of high water
5436 (Parker et al. 1955). While inflows from this historically eutrophic lake were undoubtedly
5437 enriched in P compared with the Everglades, the influence of these inputs were likely limited to
5438 wetlands along the lake's southern fringe (Snyder and Davidson 1994) as is demonstrated by the
5439 limited extent of pond apple and other vegetation that require more nutrients for growth than the
5440 sawgrass (Cladium jamaicense) that dominates most of the Everglades.
5441
5442 Interior areas of the Everglades generally retain the oligotrophic characteristics of the
5443 predrainage ecosystem and, thus, provide the best contemporary information on historical P
5444 concentrations. Water chemistry data were available for several interior locations that had been
5445 sampled by the State for many years. Median water-column TP concentrations at these stations
5446 ranged between 4 and 10 jig L"1, with lowest concentrations occurring in southern areas that
5447 have been least affected by anthropogenic P loads (Fig. 3). Phosphorus concentrations >10 jig L"
5448 * were measured periodically at many of these sites. Isolated high P concentrations at reference
5449 stations were attributed to P released as a result of oxidation of exposed soils, increased fire
5450 frequency during droughts, and difficulties in collecting water samples that are not contaminated
5451 by flocculent wetland sediments when water depths are low. Data from reference sites may
148
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
5452 represent an upper estimate of historical TP concentrations in the Everglades since several
5453 stations are located in areas that either have been overdrained, a condition which promotes soil
5454 oxidation and P release, or so heavily exposed to canal inflows (e.g., WCA 2A) that some P
5455 inputs have likely intruded even into interior areas. However, in the absence of reliable
5456 historical data, these values were deemed as best available for defining reference condition.
5457
5458 Soil Phosphorus
5459
5460 Extensive soil mapping projects across interior portions of the central and northern Everglades
5461 indicate a reference range for soil TP in the surface 0-10 cm of soil of between 200 and 500 mg
5462 kg"1 on a mass basis (DeBusk et al. 1994, Reddy et al. 1994a, Newman et al. 1997, Richardson et
5463 al. 1997a, Newman et al. 1998). Fewer data are available from ENP, but available evidence
5464 indicates background concentrations of < 400 mg kg"1 (Doren et al. 1997). Soil P content also
5465 varies volumetrically as a function of changing soil bulk density. The typical bulk density of
5466 flooded Everglades peat soils is approximately 0.08 g cm"3, whereas soils that have been
5467 subjected to extended dry out and oxidation can have bulk densities greater than 0.2 g cm"3
5468 (Newman etal. 1998). Increases in volumetric nutrient concentrations resulting from increased
5469 bulk density can have a stimulatory effect on plant growth even in the absence of external P
5470 inputs (see Chapter 2). Following correction for the varying bulk densities in the peat soils of
5471 the Everglades, a historical TP concentration of <40 jig cm"3 may be applicable for most regions
5472 (DeBusk et al. 1994, Reddy et al. 1994a, Newman et al. 1997, Newman et al. 1998, Reddy et al.
5473 1998). In the LNWR, most of the interior area has soil TP < 20 jig TP cm"3 (Newman et al.
5474 1997).
149
-------
December 2006- DRAFT
APPENDIX B: CASE STUDIES
1000
5475
5476
5477
5478
5479
5480
5481
5482
5483
5484
5485
5486
5487
5488
5489
5490
5491
5492
5493
5494
5495
5496
5497
5498
5499
O)
3.
100 A
£
i
10
O
±
LNWR WCA2A WCA3A
Hydrologic Unit
ENP
Appendix Bl. Figure Bl 3. Box plots showing surface-water P
concentrations at long-term monitoring stations in each major
hydrologic unit that illustrate the minimally impacted (i.e.,
reference) condition of the Everglades with respect to P. The
th
top, mid-line and bottom of each box represents the 75 , 50
-.th
-th
(median, and 25 percentile of data, respectively; the error bars
represent the 90th and 10th percentiles; open circles are data
outside the 90th percentile; the dashed line is the analytical limit
for TP (4 |ig I/1).
REFERENCE ECOLOGICAL CONDITIONS
The Everglades is perhaps the most intensively studied wetland in the world and, therefore, the
ecological attributes that defined the predisturbance structure and function of this ecosystem are
well understood compared with most wetlands. Clearly, not all of the valued ecological
attributes of this or any other wetland are affected directly by P enrichment. Thus, in order to
define the reference condition of the ecosystem with respect to the role of P, this assessment
focused on those processes and communities that are most sensitive to P enrichment. Based on
available information and preliminary scoping studies, 5 ecological features were selected as
biotic response variables. These features included three indicators of ecosystem structure, one
indicator of ecosystem function, and one indicator of landscape change. Structural indicators
included the periphyton community, dominant macrophyte populations, and the benthic
150
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
5500 macroinvertebrate community. Diel fluctuations in water column DO provided an important
5501 indicator of shifts in aquatic metabolism. The landscape indicator of change was the loss of
5502 open-water slough-wet prairie habitats—areas of high natural diversity and productivity.
5503
5504 Periphyton
5505
5506 Aquatic vegetation and other submerged surfaces in the oligotrophic Everglades interior are
5507 covered with periphyton, a community of algae, bacteria and other microorganisms. Periphyton
5508 accounts for a significant portion of primary productivity in sloughs and wet prairies (Wood and
5509 Maynard 1974, Browder et al. 1982, McCormick et al. 1998), and floating and attached
5510 periphyton mats provide an important habitat and food source for invertebrates and small fish
5511 (Browder et al. 1994, Rader 1994). These mats store large amounts of P (approaching 1 kg TP
5512 m"2 in some locations) and, thus, may play a critical role in maintaining low P concentrations in
5513 reference areas (McCormick et al. 1998, McCormick and Scinto 1999). Periphyton biomass and
5514 productivity peak towards the end of the wet season (August through October) and reach a
5515 minimum during the colder months of the dry season (January through March). Periphyton
5516 biomass in open-water habitats can exceed 1 kg m"2 during the wet season (Wood and Maynard
5517 1974, Browder et al. 1982, McCormick et al. 1998) when floating mats can become so dense as
5518 to cover the entire water surface. Aerobic conditions in slough-wet prairie habitats is maintained
5519 by the high productivity of this community and the capacity of dense algal mats to trap oxygen
5520 released during photosynthesis (McCormick and Laing 2003).
5521
5522 Two types of periphyton communities occur in reference areas of the Everglades. Mineral-rich
5523 waters, such as those found WCA 2A and Taylor Slough (ENP), support a periphyton
5524 assemblage dominated by a few species of calcium-precipitating cyanobacteria and diatoms,
5525 while the soft-water interior of LNWR contain a characteristic assemblage of desmid green algae
5526 and diatoms. Waters across much of the southern Everglades (WCA-3A and portions of ENP)
5527 tend to be intermediate with respect to mineral content and contain some taxa from both
5528 assemblages.
5529
5530 The chemical composition of periphyton in the oligotrophic Everglades is indicative of severe P
5531 limitation. Periphyton samples from reference areas of major hydrologic units within the
5532 Everglades are characterized by an extremely low P content (generally <0.05%) and extremely
5533 high N:P ratios (generally >60:1 w:w). This observational evidence for P limitation is supported
5534 by experimental fertilization studies that have shown that: 1) periphyton responds more strongly
5535 to P enrichment than to enrichment with other commonly limiting nutrients such as nitrogen
5536 (Scheldt et al. 1989, Vymazal et al. 1994); 2) periphyton changes in response to experimental P
5537 enrichment mimic those that occur along field nutrient gradients (McCormick and O'Dell 1996).
5538 Thus, it is well-established that periphyton is strongly P-limited in reference areas of the
5539 Everglades.
5540
5541 Dissolved Oxygen
151
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
5542
5543 Interior Everglades habitats exhibit characteristic diel fluctuations in water-column dissolved
5544 oxygen (DO), although aerobic conditions are generally maintained throughout much or all of
5545 the diel cycle (Belanger et al. 1989, McCormick et al. 1997, McCormick and Laing 2003). High
5546 daytime concentrations in open-water habitats (i.e., sloughs, wet prairies) are a product of
5547 photosynthesis by periphyton and other submerged vegetation. These habitats may serve as
5548 oxygen sources for adjacent sawgrass stands, where submerged productivity is low (Belanger et
5549 al. 1989). Oxygen concentrations decline rapidly during the night due to periphyton and
5550 sediment microbial respiration and generally fall below the 5 mg L"1 standard for Class III
5551 Florida waters (Criterion 17-302.560(21), F.A.C.). However, these diurnal excursions are
5552 characteristic of reference areas throughout the Everglades (McCormick et al. 1997) and are not
5553 considered a violation of the Class III standard (Nearhoof 1992).
5554 Vegetation
5555
5556 The vegetation communities characteristic of the pristine Everglades are dominated by species
5557 adapted to low P, seasonal patterns of wetting and drying, and periodic natural disturbances such
5558 as fire, drought and occasional freezes (Duever et al. 1994, Davis 1943, Steward and Ornes
5559 1983, Parker 1974). Major aquatic vegetation habitats in oligotrophic areas include sawgrass
5560 wetlands, wet prairies, and sloughs (Loveless 1959, Gunderson 1994). The spatial arrangement
5561 of these habitats is dynamic and controlled by environmental factors such as fire, water depth,
5562 nutrient availability and local topography (Loveless 1959).
5563
5564 Sawgrass (Cladium jamaicense) is the dominant macrophyte in the Everglades, and stands of this
5565 species compromise approximately 65 to 70% of the total vegetation cover of the Everglades
5566 (Loveless 1959). Wet prairies include a collection of low-stature, graminoid communities
5567 occurring on both peat and marl soils (Gunderson 1994). Dominant macrophyte taxa in these
5568 habitats include Rhynchospora, Panicum and Eleocharis (Loveless 1959, Craighead 1971).
5569 Sloughs are deeper water habitats that remain wet most or all of the year and are characterized
5570 by floating macrophytes such as fragrant white water lily (Nymphaea odorata), floating hearts
5571 (Nymphoides Aquaticum) and spatterdock (Nuphar advena) (Loveless 1959, Gunderson 1994).
5572 Submerged aquatic plants, primarily bladderworts (Utriculariafoliosa and U. purpurea in
5573 particular), also can be abundant in these habitats and, in the case of U. purpurea, provide a
5574 substrate for the formation of dense periphyton mats.
5575
152
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
5576 Several studies have concluded that macrophyte communities in the Everglades are P-limited.
5577 Sawgrass is adapted to the low-P conditions indicative of the pristine Everglades (Steward and
5578 Ornes 1975b, Steward and Ornes 1983). During field and greenhouse manipulations, sawgrass
5579 responded to P enrichment either by increasing the rate of growth or P uptake (Steward and
5580 Ornes 1975a, Steward and Ornes 1983, Craft et al. 1995, Miao et al. 1997, Daoust and Childers
5581 1999). Furthermore, additions of N alone had no effect on sawgrass or cattail growth under low-
5582 P conditions (Steward and Ornes 1983, Craft et al. 1995). Recent experimental evidence in the
5583 Everglades National Park (Daoust and Childers 1999) has shown that other native vegetation
5584 associations such as wet prairie communities are also limited by P.
5585
5586 Historically, cattail (Typha spp.) was one of several minor macrophyte species native to the
5587 Everglades (Davis 1943, Loveless 1959). In particular, cattail is believed to have been
5588 associated largely with areas of disturbance such as alligator holes and recent burns (Davis
5589 1994). Analyses of Everglades peat deposits reveal no evidence of cattail peat although the
5590 presence of cattail pollen indicates its presence historically in some areas (Gleason and Stone
5591 1994, Davis et al. 1994, Bartow et al. 1996). Findings such as these confirm the historical
5592 presence of cattail in the Everglades but provide no evidence for the existence of dense cattail
5593 stands covering large areas (Wood and Tanner 1990) as now occurs in the northern Everglades.
5594 In contrast, sawgrass and water lily peats have been major freshwater Everglades soils for
5595 approximately 4,000 years (McDowell et al. 1969).
5596 Macroinvertebrates
5597
5598 Aquatic invertebrates (e.g., insects, snails, and crayfish) represent a key intermediate position in
5599 energy flow through the Everglades food web as these taxa are the direct consumers of primary
5600 production and, in turn, are consumed by vertebrate predators. Invertebrates occupy several
5601 functional niches within the Everglades food web; however, most taxa are direct consumers of
5602 periphyton and/or plant detritus (e.g., Rader and Richardson 1994, McCormick et al. 2004).
5603 Rader (1994) sampled both periphyton and macrophyte habitats in this same area and, based on
5604 the proportional abundance of different functional groups, suggested that grazer (periphyton) and
5605 detrital (plant) pathways contributed equally to energy flow in low-nutrient areas of the
5606 Everglades.
5607
5608 The macroinvertebrate fauna of the Everglades is fairly diverse (approximately 200 taxa
5609 identified) and is dominated by Diptera (49 taxa), Coleoptera (48 taxa), Gastropoda (17 taxa)
5610 Odonata (14 taxa), and Oligochaeta (11 taxa) (Rader 1999). Most studies have focused on a few
5611 conspicuous species (e.g., crayfish and apple snails) considered to be of special importance to
5612 vertebrate predators, and relatively little is known about the distribution and environmental
5613 tolerances of most taxa. An assemblage of benthic microinvertebrates (meiofauna) dominated by
5614 Copepoda and Cladocera is also present in the Everglades (Loftus et al. 1986), but even less is
5615 known about the distribution and ecology of these organisms.
5616
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
5617 Invertebrates are not distributed evenly among Everglades habitats but, instead, tend to be
5618 concentrated in periphyton-rich habitats such as sloughs. In an early study, Reark (1961) noted
5619 that invertebrate densities in ENP were higher in periphyton habitats compared with sawgrass
5620 stands. Rader (1994) reported similar findings in the northern Everglades and found mean
5621 annual invertebrate densities to be more than six-fold higher in sloughs than in sawgrass stands.
5622 Invertebrate assemblages in sloughs were more species-rich and contained considerably higher
5623 densities of most dominant invertebrate groups. Functionally, slough invertebrate assemblages
5624 contained similar densities of periphyton grazers and detritivores, compared with a detritivore-
5625 dominated assemblage in sawgrass stands. Higher invertebrate densities in sloughs were
5626 attributed primarily to abundant growths of periphyton and submerged vegetation, which provide
5627 oxygen and a source of high- quality food.
5628
5629
5630 QUANTIFYING P IMPACTS
5631
5632 A targeted design (see Chapter 4 of this document) was used to quantify changes in key
5633 ecological attributes in response to P enrichment Discharges of canal waters through fixed
5634 water-control structures are the primary source of anthropogenic P for the Everglades and
5635 produce P gradients that extend several kilometers into the wetland in several locations. These
5636 gradients have existed for several decades and provided the clearest example of the long-term
5637 ecological impacts associated with P enrichment. Monitoring was conducted along gradients in
5638 different parts of the Everglades to assess ecological responses to P enrichment. Fixed sampling
5639 stations were located along the full extent of each gradient to document ecological conditions
5640 associated with increasing levels of P enrichment. Intensive monitoring was performed along
5641 gradients in two northern Everglades wetlands, WCA 2A and the LNWR. WCA 2A is a
5642 mineral-rich, slightly basic peatland and contains the most pronounced and well studied P
5643 gradient in the Everglades, whereas LNWR is a soft-water, slightly acidic peatland. These two
5644 wetlands represent the most extreme natural water chemistry conditions in the Everglades and
5645 support distinct periphyton assemblages and macrophyte populations while sharing dominant
5646 species such as sawgrass and water lily. Less intensive sampling along gradients in other parts
5647 of the Everglades (WCA 3 A and ENP) to confirm that P relationships were consistent across the
5648 wetland.
5649
5650 Chemical and biological conditions were measured at each sampling station along the two
5651 intensively sampled gradients. Repeated sampling, sometimes over several years was performed
5652 to ensure that temporal variation in each metric was considered in the final data analysis.
5653 Monthly surface-water sampling and less frequent soil sampling were performed to quantify P
5654 gradients in each area. Diel DO regimes, periphyton, and benthic macroinvertebrates were
5655 sampled quarterly when surface water was present. Macrophyte sampling included ground-
5656 based methods to document shifts in species composition and remote sensing to determine
5657 changes in landscape patterns. The hydrology of each site was characterized to determine
154
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
5658 whether P gradients were confounded with hydrologic gradients, which can also exert a strong
5659 influence on ecological patterns.
5660
5661 Numerous field experiments have been conducted to quantify ecological responses to P
5662 enrichment and to better understand how interactions between P enrichment and other factors
5663 such as hydrology may affect these responses. The design of these experiments varied in
5664 complexity with respect to size and dosing regime depending on the specific objective of each
5665 study and has included enclosed fertilizer plots (e.g., Craft et al. 1995), semi-permeable
5666 mesocosms receiving periodic P additions to achieve fixed loading rates in the form of periodic
5667 additions (e.g., McCormick and O'Dell 1996), flumes receiving semi-continuous enrichment at a
5668 fixed rate (Pan et al. 2000), and flumes receiving flow-adjusted dosing to achieve constant
5669 inflow concentrations (Childers et al. 2002). These experiments were useful in establishing the
5670 causal nature of responses to P enrichment documented along the P gradients described above.
5671
5672
5673 GRADIENT P CONCENTRATIONS
5674 Strong gradients in P concentrations were documented downstream of canal discharges into most
5675 Everglades wetlands (Fig. 4). Inflow TP concentrations in from 1996-1999 have averaged as
5676 high as 100 jig L"1 as compared with reference and pre-disturbance concentrations < 10 jig L"1.
5677 The degree and spatial extent of P enrichment varies among areas depending on the source and
5678 magnitude of inflows. The most extensive enrichment has occurred in the northern Everglades
5679 near EAA inflows, while southern areas (e.g., ENP) have been relatively less affected. The most
5680 extensive enrichment has occurred in WC A-2 A, which, unlike other areas, receives most of its
5681 water from canal discharges. Soil TP was strongly correlated with surface-water concentrations
5682 and exceeded 1500 mg kg"1 at the most enriched locations as compared with concentrations <
5683 500 mg kg"1 in reference areas. In general, this enrichment effect is limited to the surface 30 cm
5684 of soil depth (Reddy et al. 1998).
155
-------
December 2006- DRAFT
APPENDIX B: CASE STUDIES
150
O)
120 -
0.
I- 90
C
E
3
O 60
O
0)
+J
(0
30 -
5685
5686
5687
5688
5689
5690
5691
5692
5693
5694
5695
5696
5697
5698
5699
5700
5701
5702
5703
5704
5705
5706
5707
4 6 8 10 12
Distance from canal (km)
14
16
Appendix B2 Figure B2.4. Mean water-column TP
concentrations (1996-1999) at long-term monitoring stations
downstream of canal discharges in two northern Everglades
wetlands, WCA 2A (circles connected by solid line) and LNWR
(squares connected by dashed line). Error bars are + 1 SE.
ECOLOGICAL RESPONSES TO P ENRICHMENT
Periphyton
Periphyton responses to P enrichment include changes in productivity, biomass, and species
composition. Periphyton rapidly accumulates P from the water (McCormick et al. 2001, Noe et
al. 2003), and, thus, a strong relationship between P concentrations in the water and periphyton
is maintained along the P gradients (Grimshaw et al. 1993, McCormick et al. 1996). In fact,
increases in periphyton P may provide one of the earliest signals of P enrichment (e.g., Gaiser et
al. 2004). Rapid increases in periphyton photo synthetic activity and growth rates occur in
response to P enrichment (e.g., Swift and Nicholas 1987, McCormick et al. 1996, McCormick et
al. 2001). All of these responses are consistent with the P-limited nature of Everglades
periphyton.
156
-------
December 2006- DRAFT
APPENDIX B: CASE STUDIES
5708
5709
5710
5711
5712
5713
5714
5715
5716
5717
5718
5719
5720
5721
5722
5723
5724
5725
5726
5727
Paradoxically, these physiological responses are associated with sharply lower periphyton
biomass in P-enriched areas due to the loss of the abundant community of calcareous
cyanobacteria and diatoms that is indicative of mineral-rich reference areas. This community is
replaced by a eutrophic community of filamentous cyanobacteria, filamentous green algae, and
diatoms in areas having even slightly elevated P concentrations. For example, McCormick and
O'Dell (1996) found that the calcareous assemblage that existed at low water-column P
concentrations (TP = 5 to 7 jig L"1) was replaced by a filamentous green algal assemblage at
moderately elevated concentrations (TP = 10 to 28 jig L"1) and by eutrophic cyanobacteria and
diatoms species at even higher concentrations (TP = 42 to 134 jig L"1). These results are
representative of those documented by other investigators (e.g., Swift and Nicholas 1987, Pan et
al. 2000). Taxonomic changes in response to controlled P enrichment in field experiments have
been shown to be similar to those documented along field enrichment gradients (Fig. 5), thereby
providing causal evidence that changes in the periphyton assemblage were largely a product of P
enrichment (McCormick and O'Dell 1996, Pan et al. 2000).
100
Mesocosms
Transects
Loading rate
Water Column P
(M9 TP L-1)
Appendix Bl. Figure B1.5. Changes in percent biomass (as biovolume)
of major algal groups in field enclosures dosed weekly with different P
loads (left panel) and along a P enrichment gradient downstream of
157
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
5728 canal discharges (right panel) in WCA 2A. From McCormick and
5729 O'Dell 1996.
5730
5731 Community metabolism and dissolved oxygen concentrations
5732
5733 Phosphorus enrichment causes a shift in the balance between autotrophy and heterotrophy in the
5734 water column as a result of contrasting effects on periphyton productivity and microbial
5735 respiration. Rates of aquatic primary productivity (P) and respiration (R) are approximately
5736 balanced (P:R ratio =1) across the diel cycle in minimally impacted sloughs throughout the
5737 Everglades (Belanger et al., 1989; McCormick et al., 1997). In contrast, respiration rates exceed
5738 productivity by a considerable margin (P:R ratio « 1) at enriched locations. This change is
5739 related primarily to a large reduction in areal periphyton productivity as a result of shading by
5740 dense stands of cattail (Typha domingensis) that form a nearly continuous cover in the most
5741 enriched areas (McCormick and Laing, 2003). Increased cattail production also stimulates
5742 microbial respiration (e.g., sediment oxygen demand) (e.g., Belanger et al. 1989) due to an
5743 increase in the quantity and decomposability of macrophyte litter.
5744
5745 The shift towards dominance of heterotrophic processes with P enrichment, in turn, affects
5746 dissolved oxygen (DO) concentrations in enriched areas. For example, DO concentrations at an
5747 enriched site in WCA 2A rarely exceeded 2 mg L"1 compared with concentrations as high as 12
5748 mg L"1 at reference locations (McCormick et al., 1997). Depressed water-column DO
5749 concentrations have subsequently been documented in enriched areas of WCA 2A and the
5750 LNWR (Fig. 6) and confirmed in experimental P-enrichment studies (McCormick and Laing
5751 2003). Declines in DO along field P gradients were steepest within a range of water-column TP
5752 concentrations roughly between 10 and 30 jig L"1. Lower DO in enriched areas are associated
5753 with other changes including an increase in anaerobic microbial processes and a shift in
5754 invertebrate species composition toward species tolerant of low DO, described later in this study.
5755
5756 Macrophytes
5757
5758 Nutrient enrichment initially stimulates the growth of existing vegetation as evidenced by
5759 increased plant P content, photosynthesis, and biomass production as it does for periphyton.
5760 Persistent enrichment eventually produces a shift in vegetation composition toward species
5761 better adapted to rapid growth and expansion under conditions of high P availability. Two major
5762 shifts in Everglades plant communities have been documented along P gradients, including: 1)
5763 the replacement of sawgrass stands by cattail; 2) the replacement of slough-wet prairie habitat by
5764 cattail.
5765
5766 Sawgrass populations in the Everglades have life-history characteristics indicative of plants
5767 adapted to low-nutrient environments (Davis 1989, Davis 1994, Miao and Sklar 1998).
5768 Sawgrass responses to P enrichment include an increase in tissue P, plant biomass, P storage,
5769 annual leaf production and turnover rates, and seed production (e.g., Davis 1989, Craft and
5770 Richardson 1997, Miao and Sklar 1998). Cattail is characterized by a high growth rate, a short
158
-------
December 2006- DRAFT
APPENDIX B: CASE STUDIES
5771
5772
5773
5774
life cycle, high reproductive output, and other traits that confer a competitive advantage under
enriched conditions (Davis 1989, Davis 1994, Goslee and Richardson 1997, Miao and Sklar
1998).
WCA-2A
LNWR
1
o
a
c
re
4
8
Q
$
0°? °
O an
r^ 0.71 2
p < 0.001
O
O
o
o
0 0
o
0 0
Off 0
? ,
0
0 °
o
r2 = 0.314
p = 0.004
0
o
^)
£ 6
O
a .
Minimum
3 10 t
B
0
Cti
®*&Q
1^ = 0.615
p < 0.001
0
rfh oQ
r2 = 0.436
p < 0.001
0°
requency <1 mg/L(0/
yc ,
50
25
n .
I2 = 0.692
0 p < 0.001
O
o°8°0 <# o
%% ^°
0 0
o
00
o o
o o
r2 = 0.489
O p < 0.001
o
o
o
o.
eP
5775
5776
5777
5778
5779
5780
5781
5782
5783
5784
0 30 60 90 120 0 10 20 30 40
Water-column TP (Mg/L)
Appendix Bl. Figure Bl. 6. Relationship between water-column DO metrics and TP
concentration at several stations and time intervals along P gradients downstream of canal
discharges into 2 northern Everglades wetlands (see Fig. 1 for map). Total P concentrations are
mean values for all samples (n = 3 to 6) collected during the 3-month period preceding DO
measurements, which were typically collected over 3-4 diel cycles using dataloggers.
Correlation coefficients are Spearman rank coefficients based on all data in the plot.
Adapted from McCormick and Laing 2003.
159
-------
December 2006- DRAFT
APPENDIX B: CASE STUDIES
5785 Measurements and controlled enrichment experiments have shown that cattail growth rates
5786 exceed those of sawgrass under enriched conditions (Davis 1989, Newman et al. 1996, Miao and
5787 DeBusk 1999). The replacement of sawgrass by cattail in P enriched areas may be facilitated by
5788 disturbances such as flooding or severe fires that weaken or kill sawgrass plants and create
5789 openings. Consequently, sawgrass distributional patterns were not as clearly related to P
5790 gradients as were other ecological indicators of enrichment.
5791
5792 Sloughs and wet prairies appear to be particularly sensitive to replacement by cattail under P-
5793 enriched conditions, possibly due to the sparser vegetation cover in these habitats. The process
5794 of slough enrichment and replacement by cattail as shown in satellite imagery is supported by
5795 ground-based sampling methods (McCormick et al. 1999) that documented changes in slough
5796 vegetation and encroachment of these habitats by cattail in areas where soil TP concentrations
5797 averaged between 400 and 600 mg kg"1 and water-column TP in recent years averaged > 10 jig
5798 L"1. Eleocharis declined in response to increased soil P, Nymphaea was stimulated by
5799 enrichment and was dominant in slightly enriched sloughs. Increased occurrence of cattail in
5800 sloughs was associated with a decline in Nymphaea, probably as a result of increased shading of
5801 the water surface. These findings are consistent with those of Vaithiyanathan et al. (1995) who
5802 documented a decline in slough habitats along this same nutrient gradient and the loss of
5803 sensitive taxa such as Eleocharis at locations where soil TP exceeded 700 mg kg"1. As discussed
5804 by McCormick et al. (2002), loss of these open-water areas is a sensitive landscape indicator of P
5805 enrichment (Fig. 7).
80 -
5806
5807
5808
5809
5810
4 6 8 10 12
Distance from canal (km)
16
Figure Bl. 7. Changes in the percentage of open-water (i.e., sloughs, wet prairies,
or other opening caused by natural disturbance or airboats) cover at 94 locations
along a P enrichment gradient in WCA 2A as determined using aerial
160
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
5811 photography. Gray line shows the mean (+ 1 SE) water-column TP concentration
5812 (1996-1999) at 15 long-term monitoring stations along the gradient.
5813
5814 Benthic Macroinvertebrates
5815
5816 Macroinvertebrates are the most widely used biological indicator of water quality impacts, and
5817 several changes that occur in this community along P enrichment gradients in the Everglades are
5818 similar to those documented in response to eutrophication in other aquatic ecosystems. Several
5819 studies have documented an overall increase in macroinvertebrate abundance with increasing P
5820 enrichment (Rader and Richardson 1994, Trexler and Turner et al. 1999, McCormick et al.
5821 2004). However, differences in sampling methodology have apparently produced conflicting
5822 results with respect to changes in species richness and diversity. For example, Rader and
5823 Richardson (1994) documented an increase in both macroinvertebrate species richness and
5824 diversity with P enrichment in open-water (i.e., low emergent macrophyte cover) habitats and
5825 concluded that enrichment had not impacted this community. McCormick et al. (2004) however,
5826 using a landscape approach that involved habitat-weighted sampling, found little change in either
5827 species richness or diversity in response to enrichment. This latter study accounted for the
5828 decline in the cover of habitats such as sloughs and wet prairies, which contain the most diverse
5829 and abundant macroinvertebrate communities (Rader 1994). McCormick et al. (2004) also
5830 documented a pronounced shift in community composition with increasing P enrichment as taxa
5831 characteristic of the oligotrophic interior of the wetland are replaced by common pollution-
5832 tolerant taxa of oligochaetes and chironomids. These changes were indicative of habitat
5833 degradation as determined using biotic indices derived by the Florida DEP to assess stream
5834 condition based on macroinvertebrate composition (results available at
5835 http ://www. epa. gov/owow/wetlands/bawwg/case/fl2.html).
5836
5837 As for many other P-induced biological changes, the greatest change in the macroinvertebrate
5838 community occurred in response to relatively small increases in P concentration. Along field
5839 enrichment gradients, community shifts were associated with increases in water-column TP
5840 above approximately 10 ug L"1 (McCormick et al. 2004). Similarly, Qian et al. (2004)
5841 documented several shifts in community structure and function in response to long-term
5842 experimental dosing at average concentrations of approximately 10-15 ug L"1.
5843
5844 ESTABLISHING A P CRITERION
5845
5846 The FDEP was charged with reviewing and analyzing available P and ecological data collected
5847 throughout the Everglades to establish a numeric P criterion. A brief summary of this process is
5848 provided here and more detailed can be found in Payne et al (2002, 2003; both available at
5849 http://www.sfwmd.gov/sfer/previ ous_ecr.html).
5850
5851 The narrative nutrient standard for Class III Florida waters such as the Everglades states that "in
5852 no case shall nutrient concentrations of a body of water be altered so as to cause an imbalance in
5853 natural populations of aquatic flora and fauna." The FDEP approach to detecting violations of
161
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
5854 this standard with respect to surface-water P concentrations in the Everglades was to test for
5855 statistically significant departures in ecological conditions from those at reference sites, i.e.,
5856 interior sampling locations with background P concentrations. Biological and chemical data
5857 collected along anthropogenic P gradients throughout the Everglades were analyzed to determine
5858 P concentrations associated with such departures. Results showed that sampling sites with
5859 average (geometric mean) surface-water TP concentrations significantly greater than 10 ppb
5860 consistently exhibited significant departures in ecological condition from that of reference sites.
5861 A key finding supporting this concentration as the standard was the fact that multiple changes in
5862 each of the major indicator groups - periphyton, dissolved oxygen, macrophytes, and
5863 macroinvertebrates - all occurred at or near this same concentration (e.g., Payne et al. 2001).
5864
5865 Data from field and laboratory experiments conducted by various research groups provided
5866 valuable supporting information for understanding responses to P enrichment. While such
5867 experiments were not used directly to derive the P criterion, they established cause-effect
5868 relationships between P enrichment and ecological change that supported correlative
5869 relationships documented along field P gradients. For example, McCormick and O'Dell (1996)
5870 and Pan et al. (2000) showed that major shifts in periphyton species composition documented
5871 along field P gradients matched those elicited by controlled P dosing in field enrichment
5872 experiments. McCormick and Laing (2003) confirmed that controlled P enrichment produced
5873 declines in water-column DO similar to those measured along the gradients. Macroinvertebrate
5874 community changes were documented experimentally Qian et al. (2004).
5875
5876 While the criterion established a surface-water concentration of 10 ug L"1 TP as protective of
5877 native flora and fauna, the methodology used to measure compliance with the criterion needed to
5878 normalize background fluctuations in concentration. Additional analyses of P data collected
5879 over several years at reference sites was used to set both a longer-term average concentration and
5880 a shorter-term maximum concentration for each site. Based on these analyses, the FDEP
5881 concluded that annual maximum concentrations at a given sampling location should not exceed
5882 15 ug L"1 TP while long-term while 5-year average concentrations should not exceed 10 ug L"1
5883 TP. These limits would be applied to reference areas to ensure no further degradation and to
5884 areas already impacted by P enrichment to gauge the rate and extent of recovery in response to a
5885 suite of P control measures including agricultural BMPs and the construction of treatment
5886 wetlands to remove P from surface runoff prior to being discharged into the Everglades.
5887
162
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
5888 LITERATURE CITED
5889
5890 Bartow, S. M., Craft, C. B., and Richardson, C. J., Reconstructing historical changes in
5891 Everglades plant community composition using pollen distributions in peat, Journal of
5892 Lake and Reservoir Management, 12, 313, 1996.
5893
5894 Belanger, T. V., Scheldt, D. J., and Platko, J. R. II, Effects of nutrient enrichment on the Florida
5895 Everglades, Lake and Reservoir Management, 5, 101, 1989.
5896
5897 Browder, J.A., Cottrell, D., Brown, M., Newman, M., Edwards, R., Yuska, J., Browder, M., and
5898 Krakoski, J., Biomass and Primary Production of Microphytes and Macrophytes in
5899 Periphyton Habitats of the Southern Everglades, Report T-662, South Florida Research
5900 Center, Homestead, FL., 1982.
5901
5902 Browder, J. A., Gleason, P. J., and Swift, D. R., Periphyton in the Everglades: spatial variation,
5903 environmental correlates, and ecological implications, in Everglades: The Ecosystem and
5904 It's Restoration, Davis, S. M. and Ogden, J. C., Eds., St. Lucie Press, Delray Beach,
5905 Florida, 1994, 379.
5906
5907 Childers, D.L., R.D. Jones, J.C. Trexler, C. Buzzelli, S. Dailey, A.L. Edwards, E.E. Gaiser, K.
5908 Jayachandaran, A. Kenne, D. Lee, J.F. Meeder, J.H.K. Pechman, A. Renshaw, J.
5909 Richards, M. Rugge, LJ. Scinto, P. Sterling, and W. Van Gelder, 2002. Quantifying the
5910 effects of low level phosphorus enrichment on unimpacted Everglades wetlands with in
5911 situ flumes and phosphorus dosing. In Porter, J. and Porter, K. (eds). The Everglades,
5912 Florida Bay and Coral Reefs of the Florida Keys: An Ecosystem Sourcebook. CRC
5913 Press. Boca Raton, FL. Pages 127-152.
5914
5915 Cooper, S. R., and Goman, M., Historical changes in water quality and vegetation in WCA-2A
5916 as determined by paleoecological analyses, in An Integrated Approach to Wetland
5917 Ecosystem Science: The Everglades Experiments, Richardson, C. J., Ed., Springer-
5918 Verlag, New York, 2001, in press.
5919
5920 Craft, C. B., Vymazal, J., and Richardson, C. J., Response of Everglades plant communities to
5921 nitrogen and phosphorus additions, Wetlands, 15, 258, 1995.
5922
5923 Craft, C. B., and Richardson, C. J., Relationships between soil nutrients and plant species
5924 composition in Everglades peatlands, Journal of Environmental Quality, 26, 224, 1997.
5925
5926 Craighead, F. C., The Trees of South Florida Vol I: The Natural Environments and Their
5927 Succession, University of Miami Press, Coral Gables, Florida, 1971.
5928
163
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
5929 Daoust, R. J., and Childers, D. L., Controls on emergent macrophyte composition, abundance,
5930 and productivity in freshwater Everglades wetland communities, Wetlands., 19, 262,
5931 1999.
5932
5933 Davis, J. H., Jr., The Natural Features of South Florida, Especially the Vegetation, and the
5934 Everglades, Bulletin No. 25, Florida Geological Survey, 1943.
5935
5936 Davis, S. M., Sawgrass and cattail production in relation to nutrient supply in the Everglades, in
5937 Freshwater Wetlands and Wildlife, Sharitz, R. R., and Gibbons, J. W., Eds., USDOE
5938 Office of Scientific and Technical Information, Oak Ridge, Tennessee, 1989, 325.
5939
5940 Davis, S. M., Gunderson, L. H., Park, W. A., Richardson, J., and Mattson, J., Landscape
5941 dimension, composition, and function in a changing Everglades ecosystem, in
5942 Everglades: The Ecosystem and It's Restoration, Davis, S. M., and Ogden, J. C., Eds., St.
5943 Lucie Press, Delray Beach, Florida, 1994, 419.
5944
5945 DeBusk, W. F., Reddy, K. R., Koch, M. S., and Wang, Y., Spatial distribution of soil nutrients in
5946 a northern Everglades marsh - Water Conservation Area 2A, Soil Science Society of
5947 America Journal, 58, 543, 1994.
5948
5949 Doren, R. F., Armentano, T. V., Whiteaker, L. D., and Jones, R. D., Marsh vegetation patterns
5950 and soil phosphorus gradients in the Everglades ecosystem. Aquatic Botany, 56, 145,
5951 1997.
5952
5953 Duever, M. J., Meeder, J. F., Meeder, L. C., and McCollom, J. M., The climate of south Florida
5954 and its role in shaping the Everglades ecosystem, in Everglades: The Ecosystem and It's
5955 Restoration, Davis, S. M. and Ogden, J. C., Eds., St. Lucie Press, Delray Beach, Florida,
5956 1994,225.
5957
5958 Gaiser, E.E., L.J. Scinto, J.H. Richards, K. Jayachandran, D.L. Childers, J.C. Trexler, and R.D.
5959 Jones. 2004. Phosphorus in periphyton mats provides the best metric for detecting low-
5960 level P enrichment in an oligotrophic wetland. Water Research 38:507-516.
5961
5962 Gleason, P. J., Stone, P. A., Age, origin, and evolution of the Everglades peatland, in
5963 Everglades: The Ecosystem and It's Restoration, Davis, S. M. and Ogden, J. C., Eds., St.
5964 Lucie Press, Delray Beach, Florida, 1994, 149.
5965
5966 Gleason, P. J., Stone, P. A., Hallett, D., and Rosen, M., Preliminary report on the effect of
5967 agricultural runoff on the periphytic algae of Conservation Area 1, Unpublished report,
5968 Central and Southern Flood Control District, West Palm Beach, FL, 1975.
5969
164
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
5970 Goslee, S. C., and Richardson, C. J., Establishment and seedling growth of sawgrass and cattail
5971 from the Everglades, in Effects of Phosphorus and Hydroperiod Alterations on
5972 Ecosystem Structure and Function in the Everglades, Richardson, C. J., Craft, C. B.,
5973 Quails, R. G., Stevenson, R. J., Vaithiyanathan, P., Bush, M., and Zahina, J., Duke
5974 Wetland Center publication # 97-05, Report to the Everglades Agricultural Area
5975 Environmental Protection District, 1997, chap. 13.
5976
5977 Grimshaw, H. J., Wetzel, R. G., Brandenburg, M., Segerblom, M., Wenkert, L. J., Marsh, G. A.,
5978 Charnetzky, W., Haky, J. E., and Carraher, C., Shading of periphyton communities by
5979 wetland emergent macrophytes: decoupling of algal photosynthesis from microbial
5980 nutrient retention, Archive fur Hydrobiologie, 139, 17, 1997.
5981
5982 Gunderson, L. H., Vegetation of the Everglades: determinants of community composition, in
5983 Everglades: The Ecosystem and It's Restoration, Davis, S. M., and Ogden, J. C., Eds., St.
5984 Lucie Press, Delray Beach, Florida, 1994, 323.
5985
5986 Light, S.S., and J.W. Dineen. 1994. Water control in the Everglades: an historical perspective.
5987 Pages 47-84 in S.M. Davis and J.C. Ogden (eds.), Everglades: the ecosystem and its
5988 restoration. St Lucie Press, Delray Beach, FL.
5989
5990 Loftus, W. F., Chapman, J. D., and Conrow, R., Hydroperiod effects on Everglades marsh food
5991 webs, with relation to marsh restoration efforts, in Fisheries and Coastal Wetlands
5992 Research, Larson, G., and Soukup, M., Eds., Volume 6 of the Proceedings of the 1986
5993 Conference on Science in National Parks, US NFS and The George Wright Society, Ft.
5994 Collins, CO, 1986, 1.
5995
5996 Loveless, C. M., A study of the vegetation of the Florida Everglades, Ecology, 40, 1, 1959.
5997
5998 LOTAC 1972
5999
6000 McCormick, P. V., Chimney, M. J., and Swift, D. R., Diel oxygen profiles and water column
6001 community metabolism in the Florida Everglades, U. S. A., Archive fur Hydrobiologie
6002 140,117,1997.
6003
6004 McCormick, P. V., and Laing, J., Effects of increased phosphorus loading on dissolved oxygen
6005 in a subtropical wetland, the Florida Everglades. Wetlands Ecology and Management,
6006 2003.
6007
6008 McCormick, P.V., Newman, S., Miao, S. L., Reddy, K. R., Gawlik, D., Fitz, C., Fontaine, T. D.,
6009 and Marley, D., Ecological needs of the Everglades, in Everglades Interim Report, South
6010 Florida Water Management District, West Palm Beach, FL, 1999, chap 3.
6011
165
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6012 McCormick, P. V., and O'Dell, M. B., Quantifying periphyton responses to phosphorus
6013 enrichment in the Florida Everglades: a synoptic-experimental approach, Journal of the
6014 North American Benthological Society, 15,450, 1996.
6015
6016 McCormick, P. V., Rawlik, P. S., Lurding, K., Smith, E. P., and Sklar, F. H., Periphyton-water
6017 quality relationships along a nutrient gradient in the northern Everglades, Journal of the
6018 North American Benthological Society, 15,433, 1996.
6019
6020 McCormick, P. V., and Scinto, L. J., Influence of phosphorus loading on wetlands periphyton
6021 assemblages: a case study from the Everglades, in Phosphorus Biogeochemistry of
6022 Subtropical Ecosystems, Reddy, K. R., O'Connor, G. A., and Schelske, C. L., Eds.,
6023 CRC/Lewis Publishers, Boca Raton, FL, 1999, 301.
6024
6025 McCormick, P. V., Shuford, R. B. E. Ill, Backus, J. B., and Kennedy, W. C., Spatial and
6026 seasonal patterns of periphyton biomass and productivity in the northern Everglades,
6027 Florida, USA, Hydrobiologia, 362, 185, 1998.
6028
6029 McCormick, P. V., Shuford, R. B. E. Ill, and Rawlik, P. S., 2004, Changes in macroinvertebrate
6030 community structure and function along a phosphorus gradient in the Florida Everglades.
6031 Hydrobiologia 529:113-132.
6032
6033 McCormick, P.V., O'Dell, M.B., Shuford III, R.B.E., Backus, J.B. and Kennedy, W.C. 2001.
6034 Periphyton responses to experimental phosphorus enrichment in a subtropical wetland.
6035 Aquatic Botany 71:119-139.
6036
6037 McDowell, L. L., Stephens, J. C., and Stewart, E. H., Radiocarbon chronology of the Florida
6038 Everglades peat, Soil Science Society of America Proceedings, 33, 743, 1969.
6039
6040 Miao, S. L., Borer, R. E., and Sklar, F. H., Sawgrass seedling responses to transplanting and
6041 nutrient additions, Restoration Ecology, 5, 162, 1997.
6042
6043 Miao, S.L., and DeBusk, W. F., Effects of phosphorus enrichment on structure and function of
6044 sawgrass and cattail communities in Florida wetlands, in Phosphorus Biogeochemistry of
6045 Subtropical Ecosystems, Reddy, K. R., O'Connor, G. A., and Schelske, C. L., Eds.,
6046 CRC/Lewis Publishers, Boca Raton, FL, 1999, 275.
6047
6048 Miao, S. L., and Sklar, F. H., Biomass and nutrient allocation of sawgrass and cattail along a
6049 nutrient gradient in the Florida Everglades, Wetlands Ecology and Management, 5, 245,
6050 1998.
6051
166
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6052 Nearhoof, F., Nutrient-induced impacts and water quality violations in the Florida Everglades.
6053 Water Quality Technical Series 3(24), Bureau of Water Facilities Planning and
6054 Regulation, Department of Environmental Regulation, Tallahassee, FL, 1992.
6055
6056 Newman, S., Grace, J. B., and Koebel, J. W., Effects of nutrients and hydroperiod on Typha,
6057 Cladium, and Eleocharis: implications for Everglades restoration, Ecological
6058 Applications, 6, 774, 1996.
6059
6060 Newman, S., Reddy, K. R., DeBusk, W. F., Wang, Y., Shih, G., and Fisher, M. M., Spatial
6061 distribution of soil nutrients in a northern Everglades marsh: Water Conservation Area 1,
6062 Soil Science Society of 'America Journal, 61, 1275, 1997.
6063
6064 Newman, S., Schuette, J., Grace, J. B., Rutchey, K., Fontaine, T., Reddy, K. R., and Pietrucha,
6065 M., Factors influencing cattail abundance in the northern Everglades, Aquatic Botany, 60,
6066 265, 1998.
6067
6068 Noe, G.B., Scinto, L.J., Taylor, J., Childers, D.L., Jones, R.D., 2003. Phosphorus cycling and
6069 partitioning in an oligotrophic Everglades wetland ecosystem: a radioisotope tracing
6070 study. Freshwater Biology 48, 1993-2008.
6071
6072 Pan, Y., Stevenson, R. J., Vaithiyanathan, P., Slate, J., and Richardson, C. J. 2000. Changes in
6073 algal assemblages along observed and experimental phosphorus gradients in a subtropical
6074 wetland, USA. Freshwater Biology 339-353.
6075
6076 Parker, G. G., Hydrology of the pre-drainage system of the Everglades in southern Florida, in
6077 Environments of South Florida: Past and Present, Gleason, P. J., Ed., Memoir No. 2.,
6078 Miami Geological Society, Coral Gables, Florida, 1974, 18.
6079
6080 Parker, G. G., Ferguson, G. E., and Love, S. K., Water Resources of Southeastern Florida with
6081 Special Reference to the Geology and Groundwater of the Miami Area, Water Supply
6082 Paper 1255, United States Geological Survey, US Government Printing Office,
6083 Washington, D.C., 1955.
6084
6085 Payne, G., T. Bennett, and K. Weaver. 2001. Development of a Numeric Phosphorus Criterion
6086 for the Everglades Protection Area. Chapter 3 in the 2001 Everglades Consolidated
6087 Report. South Florida Water Management District, West Palm Beach, FL.
6088
6089 Payne, G., T. Bennett, and K. Weaver. 2002. Development of a Numeric Phosphorus Criterion
6090 for the Everglades Protection Area. Chapter 5 in the 2002 Everglades Consolidated
6091 Report. South Florida Water Management District, West Palm Beach, FL.
6092
167
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6093 Payne, G., K. Weaver, and T. Bennett. 2003. Development of a Numeric Phosphorus Criterion
6094 for the Everglades Protection Area. Chapter 5 in the 2003 Everglades Consolidated
6095 Report. South Florida Water Management District, West Palm Beach, FL.
6096
6097 Qian, S.S., R.S. King, and CJ. Richardson. 2003. Two statistical methods for the detection of
6098 environmental thresholds. Ecological Modeling 166:87-97.
6099
6100 Rader, R. B., Macroinvertebrates of the northern Everglades: species composition and trophic
6101 structure, Florida Scientist, 57, 22, 1994.
6102
6103 Rader, R. B., The Florida Everglades: Natural Variability, Invertebrate Diversity, and Foodweb
6104 Stability, in Freshwater Wetlands of North America: Ecology and Management, Batzer,
6105 D. P., Rader, R. B., and Wissinger, S. A., Eds., John Wiley and Sons, Inc. New York,
6106 1999,25.
6107
6108 Rader, R. B., and Richardson, C. J., Response of macroinvertebrates and small fish to nutrient
6109 enrichment in the northern Everglades, Wetlands, 14, 134, 1994.
6110
6111 Reark, J. B., Ecological investigations in the Everglades, 2nd annual report to the Superintendent
6112 of Everglades National Park, University of Miami, Coral Gables, Florida, 1961.
6113
6114 Reddy, K. R., Wang, Y., DeBusk, W. F., and Newman, S., Physico-chemical properties of soils
6115 in the water conservation area 3 (WCA-3) of the Everglades, Report to South Florida
6116 Water Management District, University of Florida, Gainesville, FL, 1994.
6117
6118 Reddy, K. R., Wang, Y., DeBusk, W. F., Fisher, M. M., and Newman, S., Forms of soils
6119 phosphorus in selected hydrologic units of Florida Everglades ecosystems, Soil Science
6120 Society of America Journal, 62, 1134, 1998.
6121
6122 Richardson, C. J., Craft, C. B., Quails, R. G., Stevenson, J., Vaithiyanathan, P., Bush, M., and
6123 Zahina, J., Effects of Phosphorus and Hydroperiod Alterations on Ecosystem Structure
6124 and Function in the Everglades, Duke Wetland Center publication # 97-05, report to the
6125 Everglades Agricultural Area Environmental Protection District, 1997a.
6126
6127 Scheldt, D. J, Flora, M. D., and Walker, D. R., Water quality management for Everglades
6128 National Park, American Water Resources Association, September Issue, 377, 1989.
6129
6130 SFWMD, Draft Surface Water Improvement and Management Plan for the Everglades,
6131 Supporting Information Document, South Florida Water Management District, West
6132 Palm Beach, FL, 1992.
6133
168
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6134 Snyder, G. H., and Davidson, J. M., Everglades agriculture - past, present, and future, in
6135 Everglades: The Ecosystem and its Restoration, Davis, S. M., and Ogden, J. C., Eds., St.
6136 Lucie Press, Delray Beach, Florida, 1994, 85.
6137
6138 Steward, K. K., and Ornes. W. H., Assessing a marsh environment for wastewater renovation,
6139 Journal of the Water Pollution Control Federation, 47, 1880, 1975a.
6140
6141 Steward, K. K., and Ornes, W. H., The autecology of sawgrass in the Florida Everglades,
6142 Ecology, 56, 162, 1975b.
6143
6144 Steward, K. K., and Ornes, W. H., Mineral nutrition of sawgrass (Cladium jamaicense Crantz) in
6145 relation to nutrient supply, Aquatic Botany, 16, 349, 1983.
6146
6147 Swift, D. R., and Nicholas, R. B., Periphyton and water quality relationships in the Everglades
6148 Water Conservation Areas, 1978-1982, Technical Publication 87-2, South Florida Water
6149 Management District, West Palm Beach, Florida, 1987.
6150
6151 Turner, A. M., Trexler, J. C., Jordan, C. F., Slack, S. J., Geddes, P., Chick, J. H., and Loftus, W.,
6152 Targeting ecosystem features for conservation: standing crops in the Everglades,
6153 Conservation Biology, 13, 898, 1999.
6154
6155 Vaithiyanathan, P., Zahina, J., and Richardson, C. J., Macrophyte species changes along the
6156 phosphorus gradient, in Effects of Phosphorus and Hydroperiod Alterations on
6157 Ecosystem Structure and Function in the Everglades, Richardson, C. J., Craft, C. B.,
6158 Quails, R. G., Stevenson, J., Vaithiyanathan, P., Bush, M., and Zahina, J., Duke Wetland
6159 Center publication # 95-05, report submitted to Everglades Agricultural Area
6160 Environmental Protection District, 1995, 273.
6161
6162 Vymazal, J., Craft, C. B., and Richardson, C. J., Periphyton response to nitrogen and phosphorus
6163 additions in the Florida Everglades, Algological Studies, 73, 75, 1994.
6164
6165 Willard, D. A., Weimer, L. M., and Riegel, W. L., Pollen assemblages as paleoenvironmental
6166 proxies in the Florida Everglades, Review ofPalaeobotany andPalynology, 113,213,
6167 2001.
6168
6169 Wood, E. J. F., and Maynard, N. G., Ecology of the micro-algae of the Florida Everglades, in
6170 Environments of South Florida: Past and Present, Gleason, P. J., Ed., Memoir No. 2.
6171 Miami Geological Society, Coral Gables, Florida, 1974, 123.
6172
6173 Wood, J. M., and Tanner, G. W., Graminoid community composition and structure within four
6174 Everglades management areas, Wetlands, 10, 127, 1990.
169
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6175 APPENDIX B. CASE STUDY 2: THE BENEFICIAL USE OF NUTRIENTS FROM
6176 TREATED WASTEWATER EFFLUENT IN LOUISIANA WETLANDS: A REVIEW12
6177 Introduction
6178 The ability of wetlands, especially natural wetlands, to perform certain water purification
6179 functions has been well established (Conner et al. 1989; Kadlec and Alvord 1989; Kemp et al.
6180 1985; Khalid et al. 1981 a and b; Knight et al. 1987; Nichols 1983; Richardson and Davis 1987;
6181 Richardson and Nichols 1985; U.S. EPA 1987, Kadlec and Knight 1996, Faulkner and
6182 Richardson 1989). Studies in the southeastern United States have shown that wetlands
6183 chemically, physically, and biologically remove pollutants, sediments and nutrients from water
6184 flowing through them (Wharton 1970; Shih and Hallett 1974; Kitchens et al. 1975; Boyt 1976;
6185 Nessel 1978; Yarbro 1979; Nessel and Bayley 1984; Yarbro et al. 1982; Tuschall et al. 1981;
6186 Kuenzler 1987). Nitrogen, in particular, undergoes numerous chemical transformations in the
6187 wetland environment (Figure 1).
6188 In some parts of the country, questions remain as to the ability of wetlands to serve as long-term
6189 storage nutrient reservoirs, but there are cypress systems in Florida that continue to remove
6190 major amounts of sewage nutrients even after 20-45 years (Boyt et al. 1977; Ewel and Bayley
6191 1978; Lemlich and Ewel 1984; Nessel and Bayley 1984). Recently, Hesse et al. (1998) showed
6192 that cypress trees at the Breaux Bridge wetlands Louisiana, which have received wastewater
6193 effluent for 50 years, had a higher growth rate than nearby trees not receiving effluent.
6194 From an ecological perspective, interest in wetlands to assimilate effluent is based on a belief
6195 that the free energies of the natural system are both capable of and efficient at driving the cycle
6196 of production, use, degradation, and reuse (Odum 1978). The basic principle underlying wetland
6197 wastewater assimilation is that the rate of application must balance the rate of decay or
6198 immobilization. The primary mechanisms by which this balance is achieved are physical settling
6199 and filtration, chemical precipitation and adsorption, and biological metabolic processes
6200 resulting in eventual burial, storage in vegetation, and denitrification (Patrick 1990; Kadlec and
6201 Alvord 1989; Conner et al. 1989). Effluent discharge generally introduces nutrients as a
6202 combination of inorganic (NOs, NH4, PC^) and organic forms. Nitrogen and phosphorus from
6203 wastewater can be
Sources
1 The Hammond Wetland Wastewater Assimilation Use Attainability Analysis (UAA), Revised April 2005. John
W. Day, Robert R. Lane, Joel Lindsey, and Jason Day. Comite Resources, Inc.
J.W. Day, Jr., Jae-Young Ko, J. Rybczyk, D. Sabins, R. Bean, G. Berthelot, C. Brantley, L. Cardoch, W. Conner,
J.N. Day, A.J. Englande, S. Feagley, E. Hyfield, R. Lane, J. Lindsey, J. Mistich, E. Reyes, and R. Twilley. 2004.
The use of wetlands in the Mississippi Delta for wastewater assimilation: a review. Ocean and Coastal
Management 47: 671-691.
170
-------
December 2006- DRAFT
APPENDIX B: CASE STUDIES
6204 Appendix B2 Figure 1. Chemical transformations of nitrogen in wetlands.
6205
6206 removed by short-term processes such as plant uptake, long-term processes such as peat and
6207 sediment accumulation, and permanently by denitrification (Hemond and Benoit 1988).
6208 Wetlands with long water residence times are best suited for BOD reduction and bacteria
6209 dieback. Many pathogenic microorganisms in sewage effluent cannot survive for long periods
6210 outside of their host organisms, and root excretions from some wetland plants can kill
6211 pathogenic bacteria (Hemond and Benoit 1988). Protozoa present in shallow waters actively
6212 feed on bacteria. The presence of vegetation can also improve the BOD purifying capacity of a
6213 wetland by trapping particulate organic matter and providing sites of attachment for
6214 decomposing bacteria.
6215 In Louisiana, discharging treated effluent into wetlands can allow for the potential enhancement
6216 and restoration of the functional attributes associated with wetlands (e.g. groundwater re-charge,
6217 flood control, biological productivity) (Kadlec and Knight 1996; Rybczyk et al. 1996; Day et al.
6218 1999, 2004). Specifically, most coastal wetlands have been hydrologically altered, and are
6219 isolated from the alluvial systems responsible for their creation (Boesch et al. 1994; Day et al.
6220 2000). This makes these wetlands especially vulnerable to the high rates of relative sea level rise
171
-------
December 2006- DRAFT
APPENDIX B: CASE STUDIES
6221
6222
(RSLR: eustatic sea level rise plus subsidence) associated with deltaic systems (Penland et al.
1988) and to predicted increases in global eustatic sea level rise (Gornitz 1982, Day et al. 2004).
6223
6224
6225
6226
6227
6228
6229
6230
6231
6232
6233
6234
6235
6236
6237
6238
6239
6240
6241
hR ' -
. .•'-'.' t
Appendix B2 Figure 2. A photograph of a typical area in the Cote Gelee wetlands near Broussard, Louisiana. The Cote Gelee wetlands are
characterized by over-drained and well-oxidized soils. This has led to a high level of soil oxidation and subsidence. Exposed roots throughout
the region suggest the soil surface has subsided by 1-2 feet. This condition could lead to a massive blow-down of the forest during a major storm
passage. Subsidence in the region has been caused by a combination of impoundment by artificial levees, which has stopped the inflow of water
and soil building materials that would normally be present during spring flooding events, and by over-engineered drainage that has led to rapid
removal of any water that does enter the region. Controlled discharges of treated wastewater to these wetlands have been shown to help reduce
subsidence and increase wetland productivity.
Wetlands have been shown to persist in the face of RSLR when vertical accretion equals or
exceeds the rate of subsidence (Baumann et al. 1984; Delaune et al. 1983; Stevenson et al. 1986).
In the past, seasonal overbank flooding of the Mississippi River deposited large amounts of
sediments into the interdistributary wetlands of the delta plain (including the Atchafalaya River
alluvial plain). Not only did these floods provide an allochthonous source of mineral sediments,
which contributed directly to vertical accretion, but also the nutrients associated with these
sediments promoted vertical accretion through increased autochthonous organic matter
production and deposition, and the formation of soil through increased root growth. This
sediment and nutrient source has been eliminated since the 1930's with the completion of levees
along the entire course of the lower Mississippi, resulting in vertical accretion deficits (RSLR >
accretion) throughout the coastal region, prolonged periods of inundation, lowered productivity,
172
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6242 marsh loss, and a lack of regeneration in forested wetlands. Primarily because of these impacts,
6243 there has been a massive loss of coastal wetlands (Day, et al. 2004; Day, et al. 2000).
6244 Contributing further to the problem of vertical accretion deficits, many wetlands in the
6245 Atchafalaya River alluvial plain have been hydrologically isolated from surrounding marshes,
6246 swamps and bayous due to an exponential increase in the construction of canals and spoil banks
6247 during the past century (Turner and Cordes 1987). In addition to impeding drainage and, in
6248 many cases, physically impounding wetlands, these spoil banks also prevent the overland flow of
6249 sediments and nutrients into cypress/tupelo forests, creating essentially ombrotrophic systems
6250 from what were naturally eutrophic or mesotrophic.
6251 The total acreage of swamp forest in Louisiana has been drastically decreased by 50% from 1956
6252 to 1990 (Barras et al. 1994). Furthermore, it has been predicted that increased rates of eustatic
6253 sea level rise and associated increase in salinity could eliminate most of the remaining forested
6254 wetlands (Delaune et al. 1987). In the wetland forests of southeastern Louisiana, Conner and
6255 Day (1988) estimated vertical accretion deficits ranging from 2.5 to 10.8 mm/yr, which leads
6256 directly to increased flooding duration, frequency and intensity. Productivity decreases observed
6257 in these wetlands may be attributed to either the direct physio-chemical effects of flooding (i.e.
6258 anoxia or toxicity due to the reduced species of S and Fe), flood related nutrient limitations (i.e.
6259 denitrification or the inhibition of mineralization), nutrient limitations due to a reduction in
6260 allocthonous nutrient supplies, lack of regeneration, or most likely, a combination of these
6261 factors (Mitsch and Gosselink 2000). For those wetlands which are not threatened by rising sea
6262 level, there is a high rate of soil subsidence caused by over drainage.
6263 Recent efforts to restore and enhance wetlands in the subsiding delta region have focused on
6264 attempts to decrease vertical accretion deficits by either physically adding sediments to wetlands
6265 or by installing sediment trapping mechanisms (i.e. sediment fences), thus increasing elevation
6266 and relieving the physio-chemical flooding stress (Boesch et al 1994; Day et al. 1992, 1999,
6267 2004). Breaux and Day (1994) proposed an alternate restoration strategy by hypothesizing that
6268 adding nutrient rich secondarily treated wastewater to hydrologically isolated and subsiding
6269 wetlands could promote vertical accretion through increased organic matter production and
6270 deposition. Their work, along with other studies, has shown that treated wastewater does
6271 stimulate productivity and accretion in wetlands (Odum et al. 1975; Mudroch and Copobianco
6272 1979; Bayley et al. 1995; Turner et al. 1976; Knight 1992; Craft and Richardson 1993; Hesse et
6273 al. 1998; Rybczyk 1997). Rybczyk et al. (2002) reported that effluent application at Thibodaux,
6274 Louisiana, increased accretion rates by a factor of three.
6275 The introduction of treated municipal wastewater into the highly perturbed forested wetlands of
6276 Louisiana may be an important step towards their ecological restoration. The nutrient
6277 component of wastewater effluent increases tree productivity (Hesse et al. 1998; Rybczyk 1996),
6278 which helps offset regional subsidence by increasing organic matter deposition enhanced organic
6279 soil formation) on the wetland surface. Increasing productivity results in greater root production
6280 which leads to organic soil formation. This action can enhance the accretion necessary to offset
173
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6281 the subsidence that contributes to wetland loss (Day, et al. 2004). The freshwater component of
6282 effluent provides a buffer for saltwater intrusion events, especially during periods of drought,
6283 which are predicted to increase in frequency in the future due to global climate change (Day, et
6284 al. 2004). These ecological benefits to wetlands are in addition to providing candidate
6285 municipalities with an economical means to meet more stringent water quality standards in the
6286 future.
6287
6288 The purpose of the Louisiana Water Control Law is to protect or enhance the quality of public
6289 water, including wetlands. Three components of the water quality standards adopted by
6290 Louisiana and approved by the EPA are; 1) beneficial water uses such as propagation offish and
6291 wildlife, 2) criteria to protect these beneficial uses, and 3) an antidegradation policy which limits
6292 the lowering of water quality. Municipalities contemplating a discharge to wetlands are required
6293 to conduct a use attainability analysis (UAA) that is submitted to the Louisiana Department of
6294 Environmental Quality as part of the permit process. A UAA describes background ecological
6295 conditions of the candidate site (hydrology, soil and water chemistry, vegetation, animal
6296 populations), analyzes the feasibility of wetland treatment, and provides preliminary engineering
6297 design and cost analyses. A number of UAA studies have been carried out to examine the effect
6298 of wetlands on effluent water quality, sediment accretion, wetland productivity, and economic
6299 savings (e.g. Day, et al. 1994; Day, et al. 1997a; Day, et al. 1997b). Various aspects of these
6300 studies have been published in the scientific literature (Breaux and Day, 1994; Blahnik and Day,
6301 2000; Rybczyk, et al. 1996) and in a number of theses and dissertations (Breaux, 1992; Hesse,
6302 1994; Westphal, 2000). The following sections briefly describe some of the beneficial
6303 environmental effects of treated wastewater discharged to wetlands as documented in case
6304 studies conducted by researchers in Louisiana in cooperation with the Louisiana Department of
6305 Environmental Quality, the US EPA, the US Army Corps of Engineers, the Louisiana Sea Grant
6306 Program, the National Coastal Resources Research and Development Institute, the Louisiana
6307 Department of Natural Resources. Local governments that have participated in these studies
6308 include the towns of Thibodaux, Breaux Bridge, Amelia, Hammond, Mandeville, St. Martinville,
6309 and Broussard; and the parishes of St. Bernard, St. Charles, and Jefferson. Cost benefit and
6310 energy savings are not discussed here, but can be found in the UAA studies and some of the
6311 references listed with this review.
6312
6313 Effects on Effluent Quality - N and P Reductions
6314
6315 Loading rates and percent nutrient reductions for several municipal discharges to wetlands sites
6316 in Louisiana are listed in Table 1 below. Zhang et al. (2000) described the effects of wastewater
6317 effluent on wetland water quality in the Point au Chene wetland for the City of Thibodaux,
6318 Louisiana. In general, the researchers found that within the immediate 231 ha zone of discharge,
6319 N and P concentrations were reduced 100% and 66% respectively from effluent inflow to
6320 outflow. In a related review, Rybczyk et al. (1998) concluded that the effluent processing could
6321 be attributed to:
6322
174
-------
December 2006- DRAFT
APPENDIX B: CASE STUDIES
6323
6324
6325
6326
6327
6328
6329
6330
6331
6332
6333
6334
6335
6336
6337
6338
6339
6340
6341
6342
6343
1. The dominant species of N in the effluent is the oxidized NOs form and not the reduced
species NH4. These naturally dystrophic wetlands denitrify NO3, resulting in a net loss of
N to the system as N2 or N2O gas;
2. Loading rates are low compared to other wetlands sites. For example, the State of
Florida has adopted regulations for wetland wastewater management that established
maximum P loading rates of 9 (gm^yr"1) for hydrologically-altered wetlands, an order of
magnitude higher than most of the Louisiana sites;
3. High rates of accretion and burial of sediments in these subsiding systems provide a
permanent sink for phosphorus.
Similar water quality improvements have been documented for the wetlands at Amelia, Breaux
Bridge, and St. Bernard (Table 1). These high reduction rates of N and P indicate that these
wetlands act as a net nutrient sink. For comparison, for many of these sites, the nutrient
concentrations are low compared to Florida's tertiary advanced wastewater treatment standards
for total N and total P, 3 and 1 mgL"1, respectively.
Appendix B2 Tablel. Loading rates and percent nutrient reductions in wastewater discharges to
forested wetlands in coastal Louisiana.
Site
Amelia8
Breaux Bridge6
St. Bernard0
Thibodaux''
Treatment
Basin (ha)
1012
1475
1536
231
Nitrogen
Loading (gm~
V1)
1.96-3.92
1.87
2
3.1
Phosphorus
Loading
(gm-y1)
0.22-0.42
0.94
0.42
0.6
Nutrient
TKN
Total P
N03-N
PO4-P
Total P
TKN
Total P
N03-N
TKN
PO4-P
Total P
Effluent discharge
concentration
2.98
0.73
0.8
1
2.9
13.6
3.29
8.7
2.9
1.9
2.46
Outlet
1
0.06
<0.1
0.2
0.3
1.4
0.23
<0.1
0.9
0.6
0.85
% Reduction
66
92
100
80
87
89.7
95
100
69
68
66
6344
6345
6346
6347
6348
6349
6350
6351
6352
6353
6354
6355
6356
All concentrations are reported as mgL
"Day, etal. 1997a
bDay, etal. 1994
c Day, etal. 1997b
11 Zhang, X. etal. 2000
Removal Pathways for N and P in Coastal Wetlands
At the Point au Chene site near Thibodaux as mentioned previously, researchers have measured
loading rates (Zhang, et al. 2000), rates of sediment accretion (Rybczyk, et al. 2002), primary
production (Rybczyk, 1997 Ph D diss.; Rybczyk, et al. 1995), rates of denitrification (Boustany,
et al. 1997; Crozier, et al. 1996), sediment nutrient concentrations (Zhang, et al. 2000), and the
physical characteristics of the soil (i.e., bulk density) (Rybczyk, et al. 2002). These works have
175
-------
December 2006- DRAFT
APPENDIX B: CASE STUDIES
6357
6358
6359
6360
6361
6362
allowed for the quantification of the loss pathways of N and P at the 231 ha site and are shown in
Table 2 below.
Appendix B2 Table 2. Estimated fate of effluent N and P entering the Point au Chene/Thibodaux
site.
A.
B.
C.
D.
E.
Storage in sediments (burial).
Calculated as the mean rate of accretion in the immediate impact zone (1. 14 cm/yr) x
mean cone, of total N (4.95 mg/g) or P (1.25 mg/g) in the upper 4 cm of soil x mean
bulk density (0.13 g/cm3) of soil in the upper 4 cm.
Storage in woody vegetation.
Calculated as mean annual increase in bole wood (285 g m^yr"1) x mean cone, of N
(0.39%) and P (0. 1 1%) in wood.8
Potential denitrification rates
Total
Loading rate
Calculated as the mean hydraulic loading rate of 6.3 x 106 L"1 day x mean N and P
effluent concentrations of 12.6 and 2.46 mg L"1 respectively x basin area (231 ha)
Total N
(g mV1)
7.3
1.1
36
44.6
12.5
Total P
(g mV1)
1.8
0.03
-
1.83
2.4
6363
6364
6365
6366
6367
6368
6369
6370
6371
6372
6373
6374
6375
6376
6377
6378
6379
6380
6381
6382
6383
6384
6385
6386
6387
6388
6389
All values used to calculate removal and loading rates were derived from data collected at the Thibodaux sites except for estimate of woody tissue
N and P.
a Concentrations of N and P in woody tissue were not measured at the Thibodaux site. Concentrations used here are means from bottomland
hardwood swamps as reported by Johnston, 1991.
Increased Sediment Accretion
As indicated earlier in this review, if coastal wetlands do not accrete vertically at a rate equal to
the RSLR (RSLR: eustatic sea level rise plus subsidence) they can become stressed and can
ultimately disappear (Day, et al. 2004). In coastal regions, especially deltas, naturally high rates
of subsidence can exceed rates of eustatic sea level rise by an order of magnitude (Penland,
1988; Emery and Aubrey, 1991). Accretion deficits (sediment accretion < RSLR) in many
coastal systems are not only the result of high rates of RSLR, but also hydrologic alterations
such as dams, dikes and levees that restrict the natural movement of nutrients and suspended
sediments into wetlands (Day, et al. 2004). In systems affected by high rates of RSLR,
hydrologic alterations, or both, treated effluents can serve as a wetland restoration or
enhancement tool, and can stimulate biomass production and enhance sediment accretion rates
(Rybczyk, et al. 2002; Reddy et al. 1993). Recently, Rybczyk et al. 2002 reported on the effects
of nutrient-rich secondarily treated effluent into the subsiding, forested wetlands at Thibodaux
and found that the effluent promoted vertical accretion through increased organic matter
production and subsequent deposition and allowed accretion to keep pace with rates of RSLR
that approached 1.23 cm yr"1 in comparison to background sediment rates averaging only 0.44 ±
0.04 cm yr"1.
Feldspar horizon marker techniques have been utilized to estimate accretion rate in sites
receiving treated effluent and in adjacent control sites, both before and after wastewater
applications (Cahoon, 1989). No significant difference between pre-effluent and control
176
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6390 accretion rates was detected, and after wastewater application began, accretion rates in the
6391 application site (1.1 cm yr"1 ) were significantly higher than accretion rates measured at the
6392 control (0.14 cm yr"1). Analysis of the sediment accretion rates (accretion rate x % organic or %
6393 mineral matter) indicated that only the rates of organic matter accumulation increased
6394 significantly after effluent application began, which the authors attributed to effluent-stimulated
6395 organic matter accretion. It could also be hypothesized that nutrient enrichment would stimulate
6396 the decomposition of organic matter, thus negating any increase in accretion due to increased
6397 organic matter accumulation, and to test this hypothesis in the same study researchers measured
6398 decomposition rates and litter nutrient dynamics in the wetland application site and in the
6399 adjacent control site, both before and after applications began. A before-and-after-control-
6400 impact (BACI) statistical analysis revealed that neither leaf-litter decomposition rates nor initial
6401 leaf-litter N and P concentrations were affected by wastewater effluent; similar analysis revealed
6402 that final N and P leaf-litter concentrations did increase in the effluent application site relative to
6403 the control after effluent was applied. Wetland elevation/sediments dynamics modeling
6404 (Rybczyk, 1998) revealed that changes in wetland elevation were much more responsive to
6405 changes in primary production than to changes in rates of decomposition and suggests that
6406 increased organic matter production and accretion would offset any increases in rates of
6407 decomposition. The model also indicated that nutrient addition alone was not sufficient to lead
6408 to long term restoration of the forested wetland and that some mineral sediment input was
6409 necessary.
6410
6411 Carbon Sequestration
6412
6413 Data to date on accretion and burial indicate that addition of nutrient-rich effluents to subsiding
6414 wetlands can substantially enhance the rate of carbon burial and sequestration. For example, in
6415 Thibodaux (Point au Chene) swamp accretion rates increased and calculated carbon burial rates
6416 increased by almost a factor of three.
6417
6418 Increased Productivity
6419
6420 While stimulating vegetative productivity with treated effluent could lead to eutrophication in
6421 some aquatic systems, many wetlands, including those in coastal Louisiana, are naturally
6422 dystrophic (Day, et al. 2004). The long-term effects of effluent discharge to coastal systems can
6423 be assessed by evaluating data from a forested wetland in Breaux Bridge, Louisiana, that has
6424 been receiving wastewater for over 50 years (Blahnik and Day, 2000; Breaux and Day, 1994).
6425 Dendrological studies (Hesse, et al. 1998; Hesse 1994) to determine long-term effects on
6426 aboveground productivity. Stem wood growth rates from 1920 to 1992 was measured at the
6427 application site and control site (no wastewater application) and an annual diameter increment
6428 ratio calculated by comparing stem wood growth from each site. Before wastewater application
6429 began (according to records between 1948 and 1953) there was significantly higher growth in the
6430 control site than at the application site. However, after the onset of effluent application, there
6431 was increased growth in the application site, resulting in statistically significant higher annual
177
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6432 diameter increment ratios. Short term studies (during 1994 -1995) at the same site had similar
6433 findings, i.e. where total production was significantly higher in a new application site as
6434 compared to the old application site. This difference was attributed to increases in stem wood
6435 biomass in the new treatment site and not leaf production. Similar results were reported for the
6436 City of Amelia, Ramos wetland site (Day, et al., 1997a; Westphal, 2000) where a year-long
6437 study on primary productivity indicated enhanced litterfall in the application sites.
6438
6439 Studies have also shown that the production of herbaceous vegetation in coastal wetlands, both
6440 emergent and floating, is also stimulated by wastewater effluent, and may contribute to sediment
6441 accretion to a greater extent than does woody vegetation (Rybczyk, 1997). Percent cover is also
6442 influenced by the seasons and warm temperatures, and can affect the type of cover (i.e.,
6443 deciduous canopy to floating aquatic vegetation).
6444
6445 Regulatory and policy considerations
6446
6447 In Louisiana, scientists, state and federal regulators, and dischargers have worked closely over
6448 the past 15 years to develop an approach to meet water quality goals in terms of discharges to
6449 subsiding wetlands. The process has allowed scientists and regulators to gain a great amount of
6450 information about characterizing these coastal wetlands and developing the appropriate criteria
6451 within the state's water quality standards to protect, monitor and assess them. In these cases, a
6452 preliminary or feasibility study (two to four months) is conducted to determine whether a
6453 discharger is a candidate for this process (to discharge to a wetland site). After the feasibility
6454 study and in consultation with state and federal regulators, if it is decided to continue with the
6455 process a year-long UAA is initiated in which: 1) the background ecological conditions of the
6456 site are described (hydrology, wetland classification, soil and water chemistry, vegetation,
6457 animal populations, and toxic materials) and analyzed; and 2) the potential impacts (along with
6458 loading rates) of the wastewater discharge are evaluated. In addition to any ecological benefits,
6459 a cost-benefit analysis is also conducted. At the conclusion of the UAA, the study results are
6460 again reviewed by standards and permit staff in the Louisiana Department of Environmental
6461 Quality. If appropriate, the beneficial Clean Water Act uses and protective criteria are
6462 recommended by the Louisiana Department of Environmental Quality for adoption into the
6463 water quality standards. The UAA then forms part of the permit application process. The permit
6464 designates effluent limits for the discharge (generally at secondary treatment levels in terms of
6465 BOD and TSS parameters) and the design loading rate (and distribution) ensures high nutrient
6466 assimilation. Disinfection is required so pathogens are not discharged to the wetlands and there
6467 should be no significant industrial use of the wastewater treatment system. After the permit is
6468 issued, the discharger constructs the project, starts discharge and initiates monitoring.
6469 Monitoring is required for the life of the permit and with annual monitoring reports.
6470
6471 Wetland monitoring requirements to assess against the recommended wetland criteria are
6472 incorporated as a part of the permit. Monitoring requirements therefore may include, but are not
6473 limited to, water stage monitoring, analysis of sediment, wetland faunal assemblages for fish and
178
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6474 macroinvertebrates, and above-ground wetland productivity (tree, grass, and/or marsh grass
6475 productivity). It should be noted that recent review of the past ten years work in the wetland
6476 UAAs indicates that faunal (benthic and nekton communities) show no clear difference between
6477 areas of effluent application and control areas and may not be appropriate as criteria in many
6478 Louisiana wetlands. Examples of wetland criteria that have been promulgated for wetland sites
6479 in Louisiana's water quality standards (Louisiana Environmental Regulatory Code, Title 33, Part
6480 IX, Subpart 1, Chapter 11, §1123, Table 3) include faunal and/or vegetative species and/or
6481 abundance, naturally occurring litter fall or stem growth, and the dominance index or stem
6482 density of bald cypress. All other general and numerical criteria not specifically revised in the
6483 standards regulations would generally apply (i.e. narratives, numerical criteria for toxics, etc.).
179
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6484 References
6485 Barras, J. A., P. E. Bourgeois, and L. R. Handley. 1994. Land Loss in Coastal Louisiana: 1956-
6486 90. National Biological Survey, National Wetlands Research Center. Open File Report
6487 94-01. 4pp. 10 color plates.
6488 Baumann, R. H., J. W. Day, and C. Miller. 1984. Mississippi, deltaic wetland survival;
6489 sedimentation versus coastal submergence. Science. 224: 1093-1095.
6490 Bayley, P. B. 1995. "Understanding large river-floodplain ecosystems." Bioscience 45 (3):
6491 153-158.
6492 Blahnik, T. and J. Day. 2000. The effects of varied hydraulic and nutrient loading rates on water
6493 quality and hydrologic distributions in a natural forested wetland. Wetlands. 20: 48-61.
6494 Boesch, D. F., M. N. Josselyn, A. J. Mehta, J. T. Morris, W. K. Nuttle, C. A. Simenstad, and D.
6495 J. Smith. 1994. Scientific assessment of coastal wetland loss, restoration and
6496 management in Louisiana. Journal of Coastal Research, Special Issue No. 20.
6497 Boustany, R. G., C. R. Crozier, J. M. Rybczyk, and R. R. Twilley. 1997. Denitrification in a
6498 south Louisiana wetland forest receiving treated sewage effluent. Wetland Ecology and
6499 Management4: 273-283.
6500 Boyt, F.L. 1976. A mixed hardwood swamp as an alternative to tertiary wastewater treatment.
6501 M.S. thesis, University of Florida, Gainesville, FL.
6502 Boyt, F. L., S. E. Bayley, and J. Zoltek, JR. 1977. Removal of nutrients from treated municipal
6503 wastewater by wetland vegetation. J. Water Pollution Control Federation 49:789-799.
6504 Breaux, A. M. 1992. The use of hydrologically altered wetlands to treat wastewater in coastal
6505 Louisiana. Ph D dissertation, Louisiana State University.
6506 Breaux, A. M. and J. W. Day, Jr. 1994. Policy Considerations for Wetland Wastewater
6507 Treatment in the Coastal Zone: A Case Study for Louisiana. Coastal Management.
6508 22:285-307.
6509 Cahoon, D. R. and R. E. Turner. 1989. Accretion and canal impacts in a rapidly subsiding
6510 wetland II. Feldspar marker horizon technique. Estuaries 12(4): 260-268.
6511 Conner, W. H. and J. W. Day, Jr. 1988. The impact of rising water levels on tree growth in
6512 Louisiana. Pages 219-224 in Hook, D. D. et al. (eds.), the Ecology and Management of
6513 Wetlands, Vol. 2: Management, Use and Value of Wetlands. Croom Helm Ltd
6514 Publishers, England.
180
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6515 Conner, W.H., J. W. Day, Jr, and J. D. Bergeron. 1989. A use attainability analysis of wetlands
6516 for receiving treated municipal and small industry wastewater: a feasibility study using
6517 baseline data from Thibodaux, LA. Center for Wetland Resources, Louisiana State
6518 University, Baton Rouge, LA. 78 p.
6519 Craft, C. B. and C. J. Richardson. 1993. Peat accretions and phosphorus accumulation along a
6520 eutrophication gradient in the northern Everglades. Biogeochemistry 22:133-156.
6521 Crozier, C. R., J. M. Rybczyk, and W. H. Patrick, Jr. 1996. Spatial gradients of dissolved nitrate
6522 and nitrous oxide in a wetland forest receiving treated sewage effluent. In: Flynn, K. ed.
6523 Proc. of the Southern Forested Wetlands Ecology and Management Conference. South
6524 Carolina: Clemson University. Pp. 78-82.
6525 Day, J. W., A. M. Breaux, I. D. Hesse and J. M. Rybczyk. 1992. Wetland wastewater treatment
6526 in the Louisiana coastal zone. Barataria-Terrebone National Estuary Program: Data
6527 Inventory Workshop Proceedings: Pages 221-238.
6528 Day, J. W., A. M. Breaux, S. Feagley, P. Kemp, and C. Courville. 1994. A use attainability
6529 analysis for long-term wastewater discharge on the Cypriere Perdue Forested Wetland at
6530 Breaux Bridge, Louisiana. Coastal Ecology Institute, Louisiana State University.
6531 Day, J. W., J. M. Rybczyk, R. I. Pratt, A. Westphal, T. Blahnick, G. Garson, and G.P. Kemp.
6532 1997a. A Use Attainability Analysis of long-term wastewater discharge on the Ramos
6533 forested wetland at Amelia, LA. Coastal Ecology Institute Report, prepared for St. Mary
6534 Parish Water and Sewer Commission #1, Town of Amelia, Louisiana.
6535 Day, J. W., Rybczyk, J. M., R. Pratt, M. Sutula, A. Westphal, T. Blahnik, P. Delgado, P. Kemp,
6536 A. J. Englande, C. Y. Hu, G. Jin, H. W. Jeng. 1997b. A Use Attainability Analysis for
6537 longterm wastewater discharge to the Poydras-Verret Wetland in St. Bernard Parish,
6538 Louisiana. Coastal Ecology Institute, Louisiana State University.
6539 Day, J., Jr., J. Rybczyk, W. Conner, P. Delgado, S. Feagley, I. Hessse, R. Pratt, A. Westphal, and
6540 X. Zhang. 1998. The use of swamp forests near Thibodaux, Louisiana for application of
6541 treated municipal wastewater: Monitoring the effects of the discharge 1992-1997.
6542 Coastal Ecology Institute, Louisiana State University, Baton Rouge, Louisiana.
6543 Day, J.W., Jr., J.M. Rybczyk, L. Cardoch, W. Conner, P. Delgado-Sanchez, R. Pratt, A.
6544 Westphal. 1999. A review of recent studies of the ecological and economical aspects of
6545 the application of secondarily treated municipal effluent to wetlands in Southern
6546 Louisiana, pp. 155-166. In: L. Rozas, J. Nyman, C. Proffitt, N. Rabalais, and R. Turner
6547 (eds.) Recent Research in Coastal Louisiana. Louisiana Sea Grant College Program,
6548 Louisiana State University, Baton Rouge.
6549 Day, J., G. Shaffer, L. Britsch, D. Reed, S. Hawes, and D. Cahoon. 2000a. Pattern and process of
181
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6550 land loss in the Mississippi delta: A spatial and temporal analysis of wetland habitat
6551 change. Estuaries. 23: 425-438.
6552
6553 Day, J. W., Jae-Young Ko, J. Rybczyk, D. Sabins, R. Bean, G. Berthelot, C. Brantley, L.
6554 Cardoch, W. Conner, J.N. Day, AJ. Englande, S. Feagley, E. Hyfield, R. Lane, J.
6555 Lindsey, J. Mistich, E. Reyes, and R. Twilley. 2004. The use of wetlands in the
6556 Mississippi Delta for wastewater assimilation: a review. Ocean and Coastal
6557 Management 47: 671-691.
6558
6559 DeLaune, R.D., R.H. Baumann, and J.G. Gosselink. 1983. Relationships among vertical
6560 accretion, coastal submergence, and erosion in a Louisiana Gulf Coast marsh. Journal of
6561 Sedimentary Petrology. 53(1):0147-0157.
6562 DeLaune, R. D., C. J. Smith, W. H. Partick, and H. H. Roberts. 1987. Rejuvenated marsh and
6563 bay-bottom accretion on rapidly subsiding coastal plain of U.S. gulf coast: a second-
6564 order effect of the emerging Atchafalaya delta. Estuarine, Coastal and Shelf Science.
6565 25:381-389.
6566 Emery, K. O. and D. G. Aubrey. 1991. Sea levels, land levels and tide gauges. New York:
6567 Springer.
6568 Ewel, K. C. and S. E. Bayley. 1978. Cypress strand receiving sewage at Waldo. Pages 750-801
6569 in H.T. Odum and K.C. Ewel, Principal Investigators. Cypress wetlands for water
6570 management, recycling, and conservation. Fourth Annual Report to National Science
6571 Foundation.
6572 Faulkner, S. P. and C. J. Richardson. 1989. Physical and chemical characteristics of freshwater
6573 wetland soils. Pages 41-72 in D.A. Hammer, ed., Constructed Wetlands for Wastewater
6574 Treatment. Lewis Publishers, Michigan.
6575 Gornitz, V., S. Lebedeff, and J. Hansen. 1982. Global sea level trend in the past century.
6576 Science 215: 1611-4.
6577 Hemond, H. F. and J. Benoit. 1988. Cumulative impacts on water quality functions of wetlands.
6578 Environmental Management 12:639-653.
6579 Hesse, I. D., J. Day. 1994. Dendroecological determination of municipal wastewater effects on
6580 Taxodium distichum (L.) rich productivity in a Louisiana swamp. M.S. Thesis, Louisiana
6581 State University.
6582 Hesse, I. D, J. Day, and T. Doyle. 1998. Long-term growth enhancement of Bald cypress
6583 {Taxodium distichum) from municipal wastewater application. Environmental
6584 Management. 22:119-127.
182
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6585 Johnston, C. A. 1991. Sediment and nutrient retention by freshwater wetlands: effects on
6586 surface water quality. Critical Reviews in Environmental Control 21: 491-565.
6587 Kadlec, R. H. and H. Alvord, Jr. 1989. Mechanisms of water quality improvement in wetland
6588 treatment systems. Pages 489-498 in D.W. Fisk, ed., Wetlands: Concerns and Successes.
6589 Proceedings sponsored by American Water Resources Association, September 17-22,
6590 1989, Tampa, Florida.
6591 Kadlec, R. H., R. L. Knight. 1996. Treatment Wetlands. Boca Raton. Lewis Publishers.
6592 Kemp, G. P., W. H. Conner, and J. W. Day, Jr. 1985. Effects of flooding on decomposition and
6593 nutrient cycling in a Louisiana swamp forest. Wetlands 5: 35-51.
6594 Khalid, R.A., R.P. Gambrell, W.H. Patrick, Jr. 198 la. An overview of the utilization of
6595 wetlands for wastewater organic carbon removal. Pages 405-423 in Progress in Wetlands
6596 Utilization and Management. Proceedings of a Symposium: 9-12 June 1981. Orlando,
6597 Florida. Sponsored by: Coordinating Council on the Restoration of the Kissimmee River
6598 Valley and Taylor Creek Nubbin Slough Basin.
6599 Khalid, R.A., I.C.R. Holford, M.N. Mixon, and W.H. Patrick. 1981b. Nitrogen and phosphorus
6600 reactions in overland flow of wastewater. U.S. Environmental Protection Agency, EPA-
6601 600/S2-81-150.
6602 Kitchens, W. F., Jr., J. M. Dean, L. H. Stevenson, and J. H. Cooper. 1975. The Santee Swamp
6603 as a nutrient sink. Pages 349-366 in F.G. Howell, J.B. Gentry, and M.H. Smith, eds.,
6604 Mineral cycling in southeastern ecosystems. Tech. Info. Center, Energy Research and
6605 Development Administration, ERDA Symposium Series 36. Available as CONF-740513
6606 from NTIS, Washington, DC.
6607 Knight, R. L, T. W. McKim, H. R. Kohl. 1987. Performance of a natural wetland treatment
6608 system for wastewater management. Journal of Water Pollution Control Federation,
6609 59(8):746-754.
6610 Knight, R. L. 1992. Natural land treatment with Carolina bays. Water Environment and
6611 Technology 4:13-16.
6612 Kuenzler, E. J. 1987. Impacts of sewage effluent on tree survival, water quality and nutrient
6613 removal in coastal plain swamps. Report No. 235 of the Water Resources Research
6614 Institute of the University of North Carolina, Chapel Hill, NC. 91 p.
6615
6616 Lemlich, S.K. and K.C. Ewel. 1984. Effects of wastewater disposal on growth rates of cypress
6617 trees. J.Environ.Qual., 13(4):602-604.
183
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6618 Mitsch, W. J. and J. G. Gosselink. 2000. The value of wetlands: importance of scale and
6619 landscape setting. Ecological Economics 35 (200) 25-33.
6620 Murdroch, J. and J. A. Capobianco. 1979. Effects of Treated Effluent on Natural Marsh. J. of
6621 Water Pollution Control Federation 51: 2243-2256.
6622 Nessel, J. K. 1978. Distribution and dynamics of organic matter and phosphorus in a sewage
6623 enriched cypress swamp. M.S. thesis, University of Florida, Gainesville, FL.
6624 Nessel, J. K. and S. E. Bayley. 1984. Distribution and dynamics of organic matter and
6625 phosphorus in a sewage-enriched cypress swamp. Pages 262-278 in K.C. Ewel and H.T.
6626 Odum, eds. Cypress Swamps, University Presses of Florida, Gainesville, FL.
6627 Nichols, D. S. 1983. Capacity of natural wetlands to remove nutrients from wastewater.
6628 Journal of Water Pollution Control Federation 55(5): 495-505.
6629 Odum, H. T. 1975. Energy quality and carrying capacity of the earth. Tropical Ecology. 16:1-8.
6630 Odum, H. T. 1978. Value of wetlands as domestic ecosystems. Pages 910-930 in H.T. Odum
6631 and K.C. Ewel, eds., Cypress Wetlands for Water Management, Recycling, and
6632 Conservation. Fourth Annual Report to National Science Foundation and the Rockefeller
6633 Foundation.
6634 Patrick, W. H., Jr. 1990. Microbial reactions of nitrogen and phosphorus in wetlands. Pages
6635 52-63 in the Utrecht Plant Ecology News Report, Utrecht, The Netherlands.
6636 Penland, S., K. E. Ramsey, R. A. McBride, J. T. Mestayer, and K. A. Westphal. 1988. Relative
6637 sea level rise and delta plain development in the Terrebonne parish region. Coastal
6638 geology technical report, Baton Rouge, Louisiana: Louisiana Geological Survey. 121 pp.
6639 Reddy, K. R., R. D. DeLaune, W. F. DeBusk, M. S. Koch. 1993. Long-term nutrient
6640 accumulation rates in the everglades. Soil Science Society of America Journal 57: 1147-
6641 1155.
6642 Richardson, C. J. and J. A. Davis. 1987. Natural and artificial ecosystems: ecological
6643 opportunities and limitations. Pages 819-854 in K.R. Reddy and W.H. Smith, eds.,
6644 Aquatic Plants for Water Treatment and Resource Recovery, Magnolia Publishing, Inc,
6645 Orlando, FL.
6646 Richardson, C. J. andD. S. Nichols. 1985. Ecological analysis of wastewater management
6647 criteria in wetland ecosystems. Pages 351-391 in P.J. Godfrey, E.R. Kaynor, and S.
6648 Pelczarski, eds. Ecological considerations in wetlands treatment of municipal
6649 wastewaters, Van Nostrand Reinhold Company, NY.
184
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6650 Rybczyk, J. M., J. W. Day, I. D. Hesse, and P. Delgado-Sanchez. 1995. The impact of
6651 Hurricane Andrew on tree mortality, litterfall, nutrient flux and water quality in a
6652 Louisiana coastal swamp forest. Journal of Coastal Research 21: 340-353.
6653 Rybczyk, J. M., J. W. Day, I. D. Hesse and P. Delgado Sanchez. 1996. An overview of forested
6654 wetland wastewater treatment projects in the Mississippi River delta region. Proceedings
6655 of the Southern Forested Wetlands Ecology and Management Conference. Kathryn Flynn
6656 (Ed.) Clemson University, South Carolina. Pages 78-82.
6657 Rybczyk, J. M. 1997. The use of secondarily treated wastewater effluent for forested wetland
6658 restoration in a subsiding coastal zone. Ph D dissertation. Louisiana State University.
6659 Rybczyk, J. M., J. C. Callaway and J. W. Day, Jr. 1998. A relative elevation model (REM) for a
6660 subsiding coastal forested wetland receiving wastewater effluent. Ecological Modeling
6661 112: 23-44.
6662 Rybczyk, J. M., J. Day, and W. Conner. 2002. The impact of wastewater effluent on accretion
6663 and decomposition in a subsiding forested wetland. Wetlands. 22(1): 18-32.
6664 Shih, S. F. and D. W. Hallett. 1974. Impact of upland marsh on water quality. Internal Report,
6665 Department of Resource Planning, Central and Southern Florida Flood Control District,
6666 West Palm Beach, FL.
6667 Stevenson, J. C., L. G. Ward and M. S. Kearney. 1986. Vertical accretion in marshes with
6668 varying rates of sea level rise, Pages 241-260. In D. Wolf (ed.) Estuarine variability.
6669 Academic Press, New York.
6670 Turner, R.E. et al. 1976. Aspects of Land-Treated Waste Application in Louisiana Wetlands.
6671 In: Proceedings of the National Symposium on Freshwater Wetlands and Sewage
6672 Effluent Disposal. Tilton D.L. editor. Pages 147-167.
6673 Turner, R. and C. Cordes. 1987. Relationship between canal and levee density and coastal land
6674 loss in Louisiana. Biological Report 85(14), U.S. Fish and Wildlife Service, Washington.
6675 Tuschall, J. R., P. L. Brezonik, and K. C. Ewel. 1981. Tertiary treatment of wastewater using
6676 flow-through wetland systems. Proc. 1981 Annual Conference of the American Society
6677 of Civil Engineers, Environmental Division.
6678 U.S. Environmental Protection Agency. 1987. Report on the Use of Wetlands for Municipal
6679 Wastewater Treatment and Disposal. Office of Water, Office of Municipal Pollution
6680 Control. Submitted to: Senator Quentin N. Burdick, Chairman of Committee on
6681 Environmental and Public Works. EPA 430/09-88-005.
185
-------
December 2006- DRAFT APPENDIX B: CASE STUDIES
6682 Wharton, C.H. 1970. The southern river swamp - a multiple use environment. School of
6683 Business Administration, Georgia State University, Atlanta, GA.
6684 Westphal, A. 2000. Effects of municipal effluent on the nutrient dynamics and productivity of a
6685 coastal forested wetland in Louisiana. M. S. thesis, Louisiana State University.
6686 Yarbro, L. A. 1979. Phosphorus cycling in the Creeping Swamp floodplain ecosystem and
6687 exports from the Creeping Swamp watershed. Ph.D. dissertation, University of North
6688 Carolina, Chapel Hill, NC. 231 p.
6689 Yarbro, L.A., E. J. Kuenzler, P.L. Mulholland, and R. P. Sniffen. 1982. The influence of swamp
6690 floodplain on exports of nitrogen and phosphorus from North Carolina coastal plain
6691 watersheds. Pages 225-241 in P.M. McCaffrey, T. Beemer, and S.E. Gatewood, eds.
6692 Proc. of the Symposium, Progress in wetlands utilization and management. Coordinating
6693 Council on Restoration of Kissimee River and Taylor Creek-Nubbin Slough Basin.
6694 Zhang, X., S. Feagley, J. Day, W. Conner, I. Hesse, J. Rybczyk, and W, Hudnall. 2000. A water
6695 chemistry assessment of wastewater remediation in a natural swamp. Journal of
6696 Environmental Quality 29: 1960-1968.
186
------- |