4.2. Wetlands

Figure 4.2
Figure 4.2 A variety of wetlands can be found along the Lower St. Johns River Basin, including salt marshes in the brackish, tidal coastal areas (left), and cypress-lined, freshwater, river swamps to the south of Jacksonville, Florida (right) (Photos: Heather P. McCarthy).

4.2.1. Description

Some of the most biologically diverse and productive systems on earth, wetlands are partially or periodically inundated with water during all or part of the year (Myers and Ewel 1990). The term wetland is broadly used to describe an area that is transitional between aquatic and terrestrial ecosystems. Within the LSJRB, these ecosystems include both coastal and freshwater wetlands. Coastal wetlands include all wetlands that are influenced by the tides within the St. Johns River watershed as it drains into the Atlantic Ocean (Stedman and Dahl 2008). The term wetland also includes non-vegetated areas like tidal sand or mud flats, intertidal zones along shorelines, intermittent ponds and oyster bars. Freshwater wetlands are typically inland, landlocked or further upstream in the Middle and Upper Basins of the St. Johns River. Wetland ecosystems described in this section are typically broken down into vegetation types based on physiognomy, or growth form of the most dominant plants: 1) forested wetlands and 2) non-forested wetlands. Forested wetlands are usually freshwater and include swampy areas that are dominated by either hardwood or coniferous trees. Non-forested wetlands can be marine, estuarine or freshwater, and include areas that are dominated by soft-stemmed grasses, rushes and sedges. Non-forested wetlands include wet prairies and mixed scrub-shrub wetlands dominated by willow and wax myrtle. The SJR represents, in Florida, one of the rivers with the highest headwater to stream length ratios, with 5.3 headwaters per km of river and a total of 886 headwaters (White and Crisman 2014). Headwater wetlands are associated with grassland/prairie, hardwood forest, and pine flatwood habitats (White and Crisman 2014).

4.2.2. Significance

Wetlands perform a number of crucial ecosystem functions including assimilation of nutrients and pollutants from upland sources. The estimated nitrogen removal of 187,765 Mt per year by SJR wetlands is valued at >$400 million per year for nitrogen and the estimated phosphorous removal of 2,390 Mt per year is valued at >$500 million per year (Craft, et al. 2015). Additionally, wetlands can help to minimize local flooding, and, thereby, reduce property loss (Brody, et al. 2007). Basins with as little as 5% lake and wetland area may have 40-60% lower flood peaks than comparable basins without such hydrologic features (Novitski 1985).  In Florida between the 1991 and 2003, 48% of permits issued were within the 100-year floodplain, suggesting potential costs for recovery (Brody, et al. 2008). Wetlands also provide nursery grounds for many commercially and recreationally important fish; refuge, nesting, and forage areas for migratory birds; shoreline stabilization; and critical habitat for a wide variety of aquatic and terrestrial wildlife (Groom, et al. 2006; Mitsch and Gosselink 2000).

4.2.3. The Science and Policy of Wetlands in the U.S.: The Past, the Present, and the Future

Since the 1970s when wetlands were recognized as valuable resources, accurately describing wetland resources and successfully mitigating for the destruction of wetlands have been ongoing pursuits in this country. During the last few decades wetland science and policy have been driven by a) calculating wetland loss, and b) determining how to compensate for the loss. The result has been adaptive management and evolving regulations.

Wetland mitigation was not initially a part of the Section 404 permitting program as outlined in the original 1972 Clean Water Act, but “was adapted from 1978 regulations issued by the Council on Environmental Quality as a way of replacing the functions of filled wetlands where permit denials were unlikely” (Hough and Robertson 2009). However, it was not until 1990 that the USACE and EPA actually defined mitigation. It was defined as a three-part, sequential process: 1) permit-seekers should first try to avoid wetlands; 2) if wetlands cannot be avoided, then permit-seekers should try to minimize impacts; and 3) if wetland impacts cannot be avoided or minimized, then permit-seekers must compensate for the losses. The Past: A Focus on Wetland Acreage

During the 1980s-1990s, assessments of wetland losses (and the mitigation required as compensation) typically focused on acres of wetlands. In 1988, President G.H. Bush pledged “no net-loss” of wetlands. This pledge was perpetuated by President Clinton in 1992, and President G.W. Bush in 2002 (Salzman and Ruhl 2005). In order to ascertain whether this goal was being achieved or not, the USFWS was mandated to produce status and trends reports using the National Wetlands Inventory data. In 1983, the first report, Status and Trends of Wetlands and Deepwater Habitats in the Conterminous United States, 1950s to 1970s, calculated a net annual loss of wetlands during this time period equivalent to 458,000 acres per year (Frayer, et al. 1983). In 1991, the second report, Status and Trends of Wetlands in the Conterminous United States, mid-1970s to mid-1980s, reported a decline in the rate of loss to 290,000 acres per year (Dahl and Johnson 1991). In 2000, the USFWS released the third report, Status and Trends of Wetlands in the Conterminous United States 1986 to 1997, which concluded the net annual loss of wetlands had further declined to 58,500 acres per year (Dahl 2000). The Present: A Focus on Wetland Functions

The last decade has also been marked by the growing popularity of wetland mitigation banking. To offset the impacts of lost wetlands caused by a permitted activity, the SJRWMD or USACE (with the consent of DEP) may allow a permit-holder to purchase compensatory mitigation credits from an approved mitigation bank per the Compensatory Mitigation Rule (USACE 2008). Wetland mitigation banks are designed to compensate for unavoidable impacts to wetlands that occur as a result of federal or state permitting processes (NRC 2001). By 2008, it was reported that mitigation banking accounted for >30% of all regulatory mitigation arising from the Section 404 permitting process (Ruhl, et al. 2008). This is not a surprise as the USACE actively supports the use of mitigation banks: “Mitigation banks are a “performance-based” form of wetland and stream replacement because, unlike in-lieu fee mitigation and permittee-responsible mitigation, the tradable aquatic resource restoration credits generated by banks are tied to demonstrated achievement of project goals. Thus, the rule establishes a preference for the use of credits from mitigation banks when appropriate credits are available” (USACE 2008).

Although more successful than previous approaches, mitigation banking has its own set of inherent problems and inadequacies. As Salzman and Ruhl 2005 explain, “different types of wetlands may be exchanged for one another; wetlands in different watersheds might be exchanged; and wetlands might be lost and restored in different time frames.” According to Salzman and Ruhl 2005, “Despite all its potential shortcomings, wetland mitigation banks certainly remain popular. Credits in Florida are now trading anywhere from $30,000-$80,000 per acre. There clearly is demand and banks are still being created to supply it.” Of course, the price that a permit-holder pays per mitigation credit varies by bank and time. For example, in October 2007, SJRWMD approved the Florida Department of Transportation (FDOT) to purchase 55 mitigation bank credits from the East Central Florida Mitigation Bank at a purchase price of $32,000 per credit with up to ten additional credits for $38,000 each for unexpected impacts (SJRWMD 2007b).

In 2006, the fourth report by the USFWS, Status and Trends of Wetlands in the Conterminous United States 1998 to 2004, calculated for the first time a net gain of wetlands in the U.S. equivalent to 32,000 acres per year (Dahl 2006). This result was publicized, celebrated, scrutinized, and criticized. The central shortfall of the USFWS analyses was that wetland functions were not considered. This shortfall was briefly addressed in a footnote in the middle of the 112-page report: “One of the most important objectives of this study was to monitor gains and losses of all wetland areas. The concept that certain kinds of wetlands with certain functions (e.g., human-constructed ponds on a golf course) should have been excluded was rejected. To discriminate on the basis of qualitative considerations would have required a much larger and more intensive qualitative assessment. The data presented do not address functional replacement with loss or gain of wetland area” (Dahl 2006). The results of the 2006 report solidified the acceptance among scientists and policymakers that the simplistic addition and subtraction of wetland acres do not produce a wholly accurate portrayal of the status of wetlands. In short, any comprehensive evaluation of the status of wetlands needs to include a thorough consideration of what types of wetlands are being lost or gained and the ecosystem functions those wetlands provide.

Toward this end, publications began to emphasize that the USFWS’s reported net gain of wetlands in the U.S. must be viewed alongside some important caveats and exceptions (CEQ 2008). For instance, some important types of wetlands were declining, although the overall net gain was positive. In 2008, USFWS and NOAA released an influential report entitled Status and Trends of Wetlands in the Coastal Watersheds of the Eastern United States 1998-2004 (Stedman and Dahl 2008). This report calculated an annual loss of coastal wetlands at a rate of 59,000 acres per year (prior to Hurricanes Katrina and Rita in 2005). The report states: “The fact that coastal watersheds were losing wetlands despite the national trend of net gains points to the need for more research on the natural and human forces behind these trends and to an expanded effort on conservation of wetlands in these coastal areas” (CEQ 2008). The report emphasizes the important functions of coastal wetlands and the need for more detailed tracking of wetland gains and losses.

The positive trends reported in the earlier report did not persist. The Status and Trends of Wetlands in the Coastal Watersheds of the Conterminous United States 2004 to 2009 states: “Wetland losses in coastal watersheds have continued to outdistance wetland gains, by an estimated 360,720 acres between 2004 and 2009 due primarily to silviculture and development…. This rate of loss increased by 25 percent since the previous reporting period of 1998 to 2004” (Dahl and Stedman 2013). The Present: A Focus on Land Use Patterns

Land use patterns in the LSJRB have remained relatively unchanged between the years 2000 and 2009 (SJRWMD 2014). The land use categories of residential, pine plantations, and wetlands each contributed >10% of total acreage (Table 4.2). Greatest decreases in percent change (>1.5%) between the two years were in the land use categories: pine flatwood, pine plantations/forest regeneration, and cabbage palm and swamp (Table 4.2).

Within 50 m of a waterway or lake in the LSJRB, 45% of total acreage was classified as wetlands/saltwater marsh and 29% of total acreage as residential in 2009 (SJRWMD 2014). Wetlands and residential use have a much larger proportion of acreage within 50 m of the river than in the entire LSJRB (Table 4.2). By comparison, the proportion of pine plantations is much greater in the LSJRB away from a water body (Table 4.2). Riverside land use categories that contributed >10% of the total 12,762 acres in Duval County were military (38%), county lands (15%), and single family residential (13%) (JCCI 2005).

Table 4.2. Comparison of selected land use categories between 2000 and 2009 the Lower St. Johns River Basin, Florida (Sources: SJRWMD 2010c; SJRWMD 2014).

LAND USE2000200920002009
Pine plantation, forest regeneration24.522.23.94.5
Swamp, cabbage palm11.99.632.329.1
(not saltwater marsh)
Saltwater marsh10.98.97.8 The Present: A Focus on Wetland Mitigation Banking

The last decade has also been marked by the growing popularity of wetland mitigation banking. To offset the impacts of lost wetlands caused by a permitted activity, the SJRWMD or USACE (with the consent of DEP) may allow a permit-holder to purchase compensatory mitigation credits from an approved mitigation bank per the Compensatory Mitigation Rule (USACE, 2008a). Wetland mitigation banks are designed to compensate for unavoidable impacts to wetlands that occur as a result of federal or state permitting processes (NRC 2001). By 2008, it was reported that mitigation banking accounted for >30% of all regulatory mitigation arising from the Section 404 permitting process (Ruhl, et al. 2008). This is not a surprise as the USACE actively supports the use of mitigation banks: “Mitigation banks are a “performance-based” form of wetland and stream replacement because, unlike in-lieu fee mitigation and permittee-responsible mitigation, the tradable aquatic resource restoration credits generated by banks are tied to demonstrated achievement of project goals. Thus, the rule establishes a preference for the use of credits from mitigation banks when appropriate credits are available” (USACE 2008).

Although more successful than previous approaches, mitigation banking has its own set of inherent problems and inadequacies. As Salzman and Ruhl 2005 explain, “different types of wetlands may be exchanged for one another; wetlands in different watersheds might be exchanged; and wetlands might be lost and restored in different time frames.” According to Salzman and Ruhl 2005, “Despite all its potential shortcomings, wetland mitigation banks certainly remain popular. Credits in Florida are now trading anywhere from $30,000-$80,000 per acre. There clearly is demand and banks are still being created to supply it.” Of course, the price that a permit-holder pays per mitigation credit varies by bank and time. For example, in October 2007, SJRWMD approved the Florida Department of Transportation (FDOT) to purchase 55 mitigation bank credits from the East Central Florida Mitigation Bank at a purchase price of $32,000 per credit with up to ten additional credits for $38,000 each for unexpected impacts (SJRWMD 2007b).

To facilitate mitigation banking within northeast Florida, the SJRWMD has delineated mitigation basins. In most cases, mitigation credits can only be purchased within the same mitigation basin as the permitted project where wetland loss is expected. The SJRWMD mitigation basins closely resemble, but do not exactly align with the USGS drainage basins. Within the LSJRB, all or part of the following SJRWMD mitigation basins can be found: Northern St. Johns River and Northern Coastal, Tolomato River and Intracoastal Nested, Sixmile and Julington Creeks Nested, Western Etonia Lakes, St. Johns River (Welaka to Bayard), and Crescent Lake (SJRWMD 2010c).

The definition and use of mitigation bank service areas are explained below according to the SJRWMD (SJRWMD 2010c):

A mitigation bank’s service area is the geographic area in which mitigation credits from the bank may be used to offset adverse impacts to wetlands and other surface waters. The service area is established in the bank’s permit. The mitigation service areas of different banks may overlap. With three exceptions, mitigation credits may only be withdrawn to offset adverse impacts of projects located in the bank’s mitigation service area. The following projects or activities are eligible to use a mitigation bank even if they are not completely located in the bank’s mitigation service area:

  1. Projects with adverse impacts partially located within the mitigation service area;
  2. Linear projects, such as roadways, transmission lines, pipelines; or
  3. Projects with total adverse impacts of less than one acre in size.

Before mitigation credits for these types of projects may be used, SJRWMD must still determine that the mitigation bank will offset the adverse impacts of the project and either that:

  1. On-site mitigation opportunities are not expected to have comparable long-term viability due to such factors as unsuitable hydrologic conditions or ecologically incompatible existing adjacent land uses; or
  2. Use of the mitigation bank would provide greater improvement in ecological value than on-site mitigation.

In the LSJRB, 18 mitigation banks are active with permits processed by the USACE, and 13 mitigation banks are active with permits processed by the SJRWMD (Tables 4.3 and 4.4; ERDC 2016; SJRWMD 2016b). These mitigation banks are typically located in rural areas with palustrine habitats. Brandy Branch was approved in 2015. In 2016, 17 mitigation banks showed permit activity (Tables 4.3 and 4.4). Permits for Natural Resource, Sunnyside, St. Johns Co., Poa Bay, Mill Creek, Little Creek Florida, Nochaway, Normandy, and Lower St. Johns Mitigation Banks are currently pending and may become active next year. One of the most-used mitigation banks managed by the company Loblolly Mitigation Preserve, is currently in dispute with the city of Jacksonville (Hong 2016). The city had bought 4,200 acres to use as mitigation for city projects and also sold credits for private development projects. City council Auditor’s Office indicated that the company managing the mitigation bank owes $4 million (Hong 2016).

Table 4.3 Wetland mitigation banks permitted by the USACE serving the LSJRB, Florida (Source: ERDC 2016).
Values in parentheses indicate credits reported in 2016 River Report, if any changes were reported.
Barberville Mitigation Bank366Palustrine emergent, palustrine forested2.813.115.963.7
Brandy Branch762Palustrine forested15.6015.6130.5
Brick Road2,945Palustrine emergent, palustrine forested62.91.764.6504
Farmton24,323Palustrine3975.2 (3986.7 )363.7 (352.2 )4338.95656.6 (5102.6)
Fish Tail Swamp5,327Palustrine forested192.8 (204.9 )40.9 (28.8 )233.7860.1
Greens Creek1,353Palustrine forested48.2 (48.4)54.1 (53.9 )102.3291.9
Highlands Ranch1,581Palustrine forested14.4 (9.1)3.2 (0)17.5 (9.1)70.37
Lake Swamp1,890Palustrine31.3 (54.7)66.2 (27.0)96.9 (81.7)215.3
Loblolly6,240Palustrine forested1587.3 (1591.1)427.5 (423.7)2014.82507.5
Longleaf3,021Palustrine emergent, palustrine forested671.0 (789.5)355.7 (237.2)1026.71026.7
Northeast Florida Saltwater Marsh92.36Estuarine intertidal, emergent70749.6
Northeast Florida Wetland386Palustrine290.1 (292.5)352.9 (350.5)643643
Peach Drive57.3Palustrine forested27.72047.647.6
Star 4950.4Palustrine forested29.3 (30.5)6.83 (0.1)36.1 (30.5)182.5
St. Marks Pond935Palustrine forested, palustrine emergent10.96.717.658.5
Sundew2,105Palustrine emergent, palustrine forested91.6 (913.7)24.6116.3 (116.1)931.4
Town Branch432Palustrine forested12.7 (19.5)6.8 (0)19.556.3
Tupelo1,525Palustrine forested489.5 (491.1)134.1 (132.5)623.6623.6 (643.9)
Table 4.4 Wetland mitigation banks permitted by SJRWMD serving the Lower St. Johns River Basin, Florida
(Sources: DEP 2013h; SJRWMD 2016b). Values in parentheses indicate credits reported in 2016 River Report, if any changes were reported.
Greens Creek4,201Forested freshwater69.0 (80.7)302.6 (254.0)405.4
Highlands Ranch Mitigation1, 575Forested freshwater71.285.6200.5
Loblolly6,247Forested freshwater, general wetlands36.9 (18.6)1389.4 (1359.6)1684
Longleaf3,020Forested freshwater33.1 (54.9)(334.6375
Nochaway4,076Forested freshwater55.8 (56.9)132.4 (58.9)459.7
Northeast Saltwater Marsh93Estuarine intertidal, emergent0.01 (5.3)7.247.7
Northeast Florida Wetland774General wetlands11.4 (11.7)395.9 (394.9)395.9 (394.9)
Star 4950Forested freshwater25.9 (60.3)137.4 (103.0)171.7
Sundew2,107Forested freshwater, herbaceous freshwater182.9 (11.4)379.0 (206.9)621.7 (698.3)
Tupelo1,524General wetlands28.5 (34.5)459.7459.7
St. Marks Pond759Forested freshwater, herbaceous freshwater29.7 (38.4)53.9 (53.9)134.8
Lower St. Johns990Forested freshwater1.2 (1.3)117.1140.1 The Future: A Focus on Wetland Services

The future of wetland policies is rising out of the emerging science of ecosystem services (Ruhl, et al. 2008). As applied to wetlands, the science of ecosystem functions investigates how wetlands function in ecosystems (e.g., as nursery grounds, shelter, or food for wildlife). The emerging science of ecosystem services examines how wetlands serve human populations. As explained by Ruhl, et al. 2008, recent research documents that “wetlands can provide important services to local populations, such as air filtering, micro-climate regulation, noise reduction, rainwater drainage, pollutant treatment, and recreational and cultural values.”

Ecosystem services research is just beginning to develop cost-effective methods to quantify wetland alterations. For example, wetland mitigation banking has led to a predominance of wetland banks in rural areas (Ruhl and Salzman 2006). In this case, the services provided by wetlands are taken from urban to rural environments. These services, like sediment capture, groundwater recharge, water filtration, and flood mitigation, have economic value associated with them. Calculating the dollar value of such services is a challenging, but not impossible, endeavor (Figure 4.3). The economic value of wetlands to retain storm water surges or buffer shorelines was clear after Hurricanes Katrina and Rita hit the Gulf Coast of the U.S., where coastal wetlands have been substantially diminished (Stedman and Dahl 2008). Brody, et al. 2007 examined wetland permits granted by the USACE in Florida between 1997 and 2001 and determined that “one wetland permit increased the average cost of each flood in Florida by $989.62.”

Figure 4.3
Figure 4.3 Estimated value of ecosystem services by habitat (Source: Brown and Shi 2014).

Likewise, the economic value of wetland-dependent recreation in northeast Florida is estimated in the range of $700 million per year (Kiker and Hodges 2002). The wetland-dependent activities with the greatest economic value to northeast Florida are recreational saltwater fishing ($301.6 million per year), followed by wildlife viewing ($226.5 million per year). Based on survey results, Florida residents and tourists value outdoor recreation (>95% of 3,961 Florida residents and 2,306 tourists participated in outdoor recreation) and specifically saltwater beach activities (63%), wildlife viewing trips (49%), and fishing (46%) (DEP 2013i). In Florida, 2.9 million people fished, hunted, or viewed wildlife in 2006 (USDOI and USDOC 2008). The number of pleasure vessels recorded in Duval, St. Johns, Clay, Putnam, and Flagler is greater than 500,000 vessels (SRR 2012). Bird watchers spent an estimated $3.1 billion and fishers $4.3 billion in 2006 (USDOI and USDOC 2008). Canoeing and kayaking have become more popular, representing 14% of recreational activities in 2002 and 26% in 2011 (DEP 2013i). If these kinds of services are negatively impacted, the economic and social repercussions can be substantial.

Partially in response to the growing body of knowledge regarding wetland services, the USACE and EPA published a landmark overhaul of U.S. wetland regulations in April 2008 (USACE 2008). Not only did the rule consolidate the regulatory framework and require consideration of wetland functions, according to Ruhl, et al. 2008, “the new rule also for the first time introduces ecosystem services into the mitigation decision-making standards, requiring that ‘compensatory mitigation…should be located where it is most likely to successfully replace lost…services.’” However, this requirement may be slightly ahead of the science – the necessary databases and scientific methods needed to fully consider the costs and benefits of ecosystem services do not yet exist. Although the new rule acknowledges that compensatory mitigation affects how wetland services are distributed and delivered to distinct human populations, there are few methods available for assessing these services quickly and reliably at any given site.

4.2.4. Data Sources on Wetlands in the LSJRB Data Sources for Wetland Spatial Analyses

Nine GIS (Geographic Information System) maps that contain data on wetlands vegetation were available and analyzed. The GIS maps were created by either the Department of Interior USFWS or the SJRWMD from high-altitude aerial photographs (color infrared or black-and-white photos) with varying degrees of consideration of soil type, topographical and hydrologic features, and ground-truthing. In this analysis, each parcel of land or water was outlined and assigned a category, creating distinct polygons for which area (i.e., number of acres) can be calculated. These areas were used to calculate total wetlands and total acres within the LSJRB for each year available (Table 4.5).

Table 4.5 Comparison of wetland maps – Lower St. Johns River Basin, Florida.

SJRWMD-corrected National Wetlands Inventory map (produced from 1971-1992 lumped data, processed by SJRWMD in 2001, 2003)727,631849,512 ACRES INCLUDING DEEPWATER. Non-wetland upland acres not specified in this map.
SJRWMD Wetland & Deep Water Habitats map (based on National Wetlands Reconnaissance Survey maps from 1972-1980, processed 1996 by SJRWMD, dated 2001)870,5763,110,209
SJRWMD Wetlands & Vegetation Inventory map (based on District's Wetlands Mapping Project 1984-2002, finished 2002, accuracy of wetland boundaries estimated at 80-95%)441,0722,208,172
SJRWMD Land Use/Land Cover map (based on 1973 data)440,0482,100,552
SJRWMD Land Use/Land Cover map (based on 1990 data)435,6622,605,247
SJRWMD Land Use/Land Cover map (based on 1995 data)450,5951,910,422
SJRWMD Land Use/Land Cover map (based on 2000 data)444,4671,851,447
SJRWMD Land Use/Land Cover map (based on 2004 data)455,3081,868,003
SJRWMD Land Use/Land Cover map (based on 2009 data)452,3151,903,789
* Lumped dates for maps result from the consolidation of aerial photographs taken during different years.* 1.8 million acres is considered the accurate area of the LSJRB (according to the SJRWMD). Demonstrates that maps are not statistically comparable for total wetland area. Data Sources for Wetland Permit Analyses

Within the LSJRB, there are two governmental entities that grant permits for the destruction, alteration, and mitigation of wetlands: 1) SJRWMD, and 2) U.S. Army Corps of Engineers (USACE). The differing regulatory definitions of wetlands used by Federal and State agencies are outlined in Appendix 4.2.A. At the regional level, the SJRWMD has posted a comprehensive online database of all mitigation bank ledgers (SJRWMD 2010c). At the national level, the USACE and EPA have made available a single online database to track mitigation banking activities called the Regional Internet Bank Information Tracking System (RIBITS) (ERDC 2015). Concurrently, the EPA and USACE have developed a GIS-enabled database to spatially track and map permits and mitigation bank transactions, which will interface and complement the RIBITS database (Ruhl, et al. 2008).

The wetland permit analysis conducted for this report reveals how the acreage of wetlands has changed over time according to the historical wetland permits granted through the SJRWMD Environmental Resource Permitting Program.

4.2.5. Limitations Limitations of Wetland Spatial Analyses

The identification of vegetation type from an aerial photograph is an imperfect process. The metadata associated with the SJRWMD Wetlands & Vegetation Inventory map estimates the margin of error in wetlands delineation from aerial photographs to vary according to the type of vegetation being identified and range from 5-20% (SJRWMD 2010b). The metadata states: “The main source of positional error, in general, is due to the difficulty of delineating wetland boundaries in transitional areas. Thematic accuracy: correct differentiation of wetlands from uplands: 95%; correct differentiation of saline wetlands from freshwater or transitional wetlands: 95%; correct differentiation of forested, shrub, herbaceous, or other group forms: 90%; correct differentiation of specific types within classes: 80%. Accuracy varies for different locations, dates, and interpreters.”

In addition to interpretational errors, wetland maps do not accurately reflect wetlands habitats that vary seasonally or annually (e.g., the spatial extent of floating vegetation or cleared areas can be dramatically different depending on the day the aerial photo was taken). Aerial photographs pieced together to create wetlands maps may be of different types (high altitude vs. low altitude, color infrared, black-and-white, varying resolutions, and varying dates). Sometimes satellite imagery is used to create wetlands maps, which is considered less accurate for wetland identification (USGS 1992).

Analyses are further limited by inconsistencies and shortcomings in the wetland classification codes used (e.g., wetland codes used in the SJRWMD Land Use/Land Cover map of 1973 were markedly different than codes used since 1990). Additionally, wetland classification codes do not always address whether a wetland area has been diked/impounded, partially drained/ditched, excavated, or if the vegetation is dead (although the National Wetlands Inventory adds code modifiers to address the impacts of man). Further, wetland mapping classification categories often do not differentiate between natural and manmade wetlands. For example, naturally occurring freshwater ponds may be coded identically with ponds created for stormwater retention, golf courses, fishing, aesthetics, water management, or aquaculture. Some maps classify drained or farmed wetlands as uplands, while others classify them as wetlands. An unknown number of additional discrepancies may exist between maps. Lastly, most of the spatial information in wetlands maps has not been ground-truthed or verified in the field but is based on analyses of aerial photographs and other maps. Limitations of Wetland Permit Analyses

A shortcoming of the records of wetlands impacted through regulatory permitting processes is that they do not address total wetland acres in the region. Additionally, acreages recorded as mitigated wetlands do not always represent an actual gain of new wetland acres (e.g., mitigation acres may represent preexisting wetlands in a mitigation bank or formerly existing wetland acres that are restored or enhanced). Thus, a true net change in wetlands (annually or cumulatively) cannot be calculated from permit numbers with certainty.

Further, changing environmental conditions require that field verification of mitigated wetlands occur on a regular basis over long time periods. The actual spatial extent, functional success, health of vegetation, saturation of soil, water flow, etc. of mitigated wetlands can change over time. On-ground site visits can verify that the spatial extent of anticipated wetlands impacted (as recorded on permits) equals actual wetlands impacted and confirm the ecological functionality of mitigated wetlands.

4.2.6. Current Status

The current status of wetlands in Florida is considered UNSATISFACTORY, because a historical decrease in wetlands has been documented statewide. The current status of wetlands in the LSJRB is considered UNSATISFACTORY because of the continued stressors to wetlands. Current Status of Wetlands in the LSJRB

The conclusions on the current status of total wetland acreage in the LSJRB that can be gleaned from GIS maps are limited. Total wetland acres in the LSJRB cannot be determined with certainty from available data. The high margin of error associated with the delineation of wetlands from aerial photographs renders the wetlands maps unsuitable for total acreage calculations (see differences in total wetlands areas and total land/water areas calculated from maps listed in Table 4.5). However, assessment of wetland loss at project sites is feasible and is a routine part of the permitting process.

Based on the SJRWMD-corrected National Wetlands Inventory Map (thought to be most accurate and complete for this kind of information), 83% of all wetlands in the LSJRB are freshwater, and 3% are estuarine and marine wetlands (Figure 4.4). Freshwater wetlands are dominated mostly by freshwater forests, followed by freshwater unconsolidated bottoms and shores (ponds). According to National Wetlands Inventory (USFWS 1984), Clay Co. and had approximately 59,771 acres, Putnam Co. had 111,198 acres, Duval Co. had 123,867 acres, Flagler Co. had 108,009, and St. Johns Co. had 114,693 acres of vegetated wetland.

Federal, state, local, and privately managed wetlands can be found in Florida conservation lands that include national parks, state forests, preserves and parks, wildlife management areas, mitigation banks, and conservation easements. For example, the North Florida Land Trust has acquired for preservation of >12,000 acres this past year (NFLT 2016). Many of these conservation lands have swamps, marshes, and other types of wetlands. Of the 895,517 acres that are managed within the LSJRB, 49% are federal lands, 45% state lands, 3% city lands, and 3% private as of December 2016 (FNAI 2016). Stat lands increased the most by 163,530 acres relative to last year, with >58,000 acres set aside as conservation easements. From a study of 20 conserved natural areas in Florida, ecosystem services were valued at $5,052 per acre (Brown and Shi 2014). For example, Pumpkin Hill Creek Preserve State Park was estimated in providing $6,169 per acre (Brown and Shi 2014).

Figure 4.4
Figure 4.4 The percentages of each wetland type in the Lower St. Johns River Basin, Florida (Source: SJRWMD 2010b).

Stressors to wetland communities include land use, nutrients, pollutants, and invasive species. In addition changes in populations of endangered/sensitive species can be indicators of stressed wetlands. Below is a discussion of these stressors affecting the LSJRB:

LAND USE. Land use is a powerful predictor of wetland condition (Reiss and Brown 2007). In Florida, countless non-tidal wetlands <5 ha that were formerly in agricultural fields and pasture lands have since been developed for residential and commercial uses (Reiss and Brown 2007). For example, in 1960, the population density was 43 people/km2 as compared to 183 people/km2 in 2000 near Deland, FL (Weston 2014). Landscape Development Intensity (LDI) is an index that associates nonrenewable energy use (electricity, fuels, fertilizers, pesticides, irrigation) to wetland condition.  Palustrine wetlands surrounded by multi-family residential, high-intensity commercial, and central business district had LDI scores of 9.19 to 10.00 as compared to pine plantation, recreational open space (low intensity) and pastures of 1.58 to 4.00 (Reiss and Brown 2007).

High LDI values can be predicted for areas in the LSJRB with multi-family residential and commercial land use. Residential land is prevalent along waterways, representing 29% of total acreage within 50 m of a waterway (Table 4.2). Surface drainage basins with residential land use can be plagued by low oxygen (e.g., Hogan Creek) and fecal coliform (e.g., Cedar River, Ginhouse Creek, Hogan Creek, Goodbys Creek, Moncrief Creek, Black Creek, Pottsburgh Creek, and Broward River) (SRR 2014). Leaking septic tanks, stormwater runoff, and wastewater treatment plants contribute to fecal coliform. Commercial activities also ranked with high LDI values (Reiss and Brown 2007). In the LSJRB, Georgia-Pacific, power plants, shipping and maritime activities, and the U.S. Department of Defense contribute to PAH, PCB, mercury, and nitrates in Rice Creek, Cedar River, and Ortega River (SRR 2016). Additional sources of PCB contamination are from waste oil spills and accidental release of locomotive waste, such as hydraulics and lubricants into drainage ditches (Flowe 2016).

The extent of the surface drainage basin can exacerbate land use pressures (e.g., stormwater runoff). For example, the surface drainage basin Etonia Creek that includes the polluted Rice Creek covers 355 miles2 (Bergman 1992). Connected surface drainage basins with a history of elevated fecal coliform levels and low oxygen include Julington Creek, Sixmile Creek, and Arlington River, covering approx. 260 miles2 (Bergman 1992). Agriculture, although with a lower LDI (Reiss and Brown 2007) can contribute to nitrogen and phosphorous loading as is recorded from Deep Creek and Dunns Creek and cover approx. 100 miles2 of surface drainage basin (Bergman 1992; SRR 2014).

NUTRIENTS. Stormwater runoff from residential and agricultural land use can contribute more nitrogen and phosphorous than other land use categories. For example, residential areas can release 2.32 mg N/L and 0.52 mg P/L as compared to agriculture (3.47 mg N/L and 0.61 mg P/L, respectively) and undeveloped/rangeland/forest (1.15 mg N/L and 0.055 mg P/L), respectively (Harper and Baker 2007). From a 2003-2009 study of water quality collected from 59 groundwater wells in the LSJRB, a relationship was evident between land use and groundwater (Ouyang, et al. 2012).  From the shallow groundwater system, septic tank land use had greater values of nitrate/nitrite concentrations than in agricultural lands (7.4 mg/L nitrate/nitrite and 0.04 mg/L, respectively). By comparison, calcium, sodium, chlorine, and sulfate had more than twice the values in agricultural lands (agriculture: 85.9, 148.8, 318.8, and 233.1 mg/L; septic tank land use: 34.5, 23.2, 36.5, and 58.8 mg/L, respectively; Ouyang, et al. 2014). Managed plantations that use nitrogen and phosphorous excessively can be a source of nutrient loading to nearby tributaries that can be measured from weeks to years following application in sediments and water column (Shepard 1994). In addition, nutrient-laden waters from wastewater treatment spray fields can travel via the aquifer and contribute to nutrient loading far from the source, as has been recorded in Wakulla Spring from Tallahassee’s wastewater reuse facility (Kincaid, et al. 2012).

Nitrogen values remain high in a number of tributaries and sections of the LSJR (SRR 2016). By comparison, total phosphorous concentrations in the marine/estuarine river region were 33% lower in 2014 relative to 1997. However, estimates from the freshwater sections of the river increased by 45% (SRR 2016).

Sediments retain nitrogen and phosphorous. During periods of anoxic conditions due to algal blooms, Malecki et al. reported that 21% of total P load and 28% of total N load came from the sediments in the LSJR (Malecki, et al. 2004). Dissolved reactive phosphorous released from the sediments was 37 times lower (0.13 mg per m2 per day) than during aerobic conditions (4.77 mg per m2 per day) (Malecki, et al. 2004).

The presence of nutrients in combination with herbicides such as atrazine has been shown to have negative impacts on the native Vallisneria americana (Dantin, et al. 2010). Submerged aquatic vegetation (SAV; e.g., Vallisneria americana) provides food and refuge for shrimp, blue crabs, and a variety of other fauna.

POLLUTANTS. Arsenic is present in LSJRB sediments. In Naval Station Mayport, spoils from dredging of the basin were used to fill in wetlands and low-lying areas (Fears 2010). These dredged materials are concentrated in arsenic (Fears 2010). Arsenic contamination has also been documented in golf course soils (5.3 to 250 ppm, with an average of 69.2 ppm) due to herbicide applications to turf grass (81 golf courses from the northeast, 1086 surveyed in Florida; Ma, et al. 2000). Leaching of arsenic is further exacerbated by the presence of phosphorous, commonly applied in fertilizer. Many of these golf courses have waterbodies or are near wetlands, streams, and rivers (Ma, et al. 2000). Ouyang, et al. 2014 reported greater arsenic values in the groundwater associated with agriculture (4.3 µg/L) and wastewater sprayfield (5.6 µg/L) land use as compared to undeveloped forest lands and septic tank land use (0.6 and 1.3 µg/L, respectively) in the LSJB.

Wading birds and other fauna that forage in wetlands are at risk of bioaccumulation of heavy metals. For example, mercury has been reported in the Broward and Trout Rivers. Ouyang, et al. 2012 estimated an average annual mercury load of 0.36 g ha-1 year-1 within the Cedar and Ortega watershed (254 km2). St. Johns River Power Park and Northside Generating Station have reduced their mercury atmospheric emissions by 71% between 2001 and 2013 (SRR 2016). However, an increase of 250% in metal discharge was reported for electric utility since 2001, in particular zinc, nickel, cobalt, and manganese (SRR 2016).  Salt marshes are sinks for metals (Leendertse, et al. 1996). Giblin, et al. 1980 found that metals in Spartina alterniflora detritus were taken up by fiddler crabs, and metals can be concentrated in bivalves near contaminated sites (Leendertse, et al. 1996). Burger, et al. 1993 reported mean lead concentrations of 3,640 ppb dry weight in young wood storks from Dee Dot colony, demonstrating the availability of lead contamination and bioaccumulation from prey items.

HYDROLOGIC MANIPULATION. Many of the mitigation banks in the LSJRB were formerly pine plantations. Hydrology in forest plantations is typically modified to minimize surface waters (Shepard 1994) that can then impact non-tidal wetland diversity and sediment and nutrient loading to nearby waterways. Erosion in plantations adds to suspended sediments in drainage waters and connecting waterways (Shepard 1994). In lowland forested habitats, storm water is retained in the forest and runoff occurs after the groundwater table reaches the surface (Sun, et al. 2000). When trees are harvested, the groundwater table rises particularly during dry periods, a phenomenon that can continue over a period of years (Sun, et al. 2000).  The decrease in evapotranspiration rates with the loss of trees is responsible for this rise in the water table (Shepard 1994).

Bernardes, et al. 2014 raised the issue of water withdrawal affecting wetlands in northeastern Florida. Depressional wetlands are typically relict sinkholes. The Florida aquifer system is crisscrossed with fractures along which groundwater can travel. Mine pits create ponds where aquifer and groundwater accumulates and thus deprives other areas of water for recharging and supporting vegetation. Where mining-related withdrawal has occurred, wetlands in nearby mitigation banks and conservation areas have dried out with the potential of becoming sinkholes. For example, the DuPont Trail Ridge Mine is in close proximity to many of the mitigation banks listed in Table 4.3 and conservation areas (e.g., Camp Blanding, Cecil Field) that wetland permitees use to mitigate wetland alteration. Water quality, hydroperiods, and water availability would be impacted (Bernardes, et al. 2014).

INVASIVE SPECIES. The most damaging invasive plant species have the capacity to do one or more of the following: reproduce and spread successfully, compete successfully against native species, proliferate due to the absence of herbivore or pathogen that can limit their populations, and alter a habitat (Gordon 1998). Invasive species can modify a wetland habitat by changing geomorphology (erosion, soil elevation, water channel), hydrology (water table depth, surface flow), biogeochemical cycling (nutrient pathways, water chemistry, nitrogen fixation), and disturbance regime. Eichhornia crassipes and Pistia stratiotes are reported to impact siltation rates, Panicum repens stabilizes edges of waterways, Hydrilla verticillata slows water flow where abundant, and E. crassipes, P. stratiotes, and H. verticillata alter water chemistry (dissolved oxygen, pH, phosphorous, carbon dioxide, turbidity, and water color) (Gordon 1998).

Where invasive plant species are dominant, native weedy species typically proliferate (Gordon 1998). In a 2002 study of 118 depressional non-tidal wetlands in Florida, macrophyte diversity and the percentage of native perennial species in urban environments were lower than in locations away from urban environments (Reiss 2006). Species that were considered the most tolerant to disturbance intensity in depressional marshes included Alternanthera philoxeroides, Cynodon dactylon, Mikania scandens, Panicum repens, and Schinus terebinthifoilius (Cohen, et al. 2004). From a survey of 74 non-tidal depressional wetlands in Florida, greater plant species richness was associated with more disturbed sites and fewer species in undisturbed and oligotrophic conditions (Murray-Hudson, et al. 2012). Ruderal or weedy species are likely to tolerate changes in the wetland-upland boundary and variability in soil saturation and water depth and extent. The authors also showed that the outer zone adjacent to the upland border of a depressional wetland with high numbers of exotics would also have a high number of exotics throughout the wetland. This pattern was true for sensitive species as well, indicating that the condition of the wetland could be predicted by the richness of suites of species along the outer band of the wetland (Murray-Hudson, et al. 2012).

ENDANGERED/SENSITIVE SPECIES. Urbanization, habitat encroachment and increased recreational activities can negatively impact breeding populations of amphibians, reptiles, and birds. Development that alters and/or fills headwaters and streams negatively impacts habitat connectivity for many stream and wetland-dependent organism in the SJR watershed (White and Crisman 2014). Animals that require a variety of wetland types would be negatively impacted by chemical pollutants and turbidity that limits prey availability.  Sensitive species associated with wetlands include the Striped newt (Notophthalmus perstriatus) that is listed as a candidate species for protection; and the flowering plants Chapman rhododendron (Rhododendron chapmanii), Okeechobee gourd (Cucurbita okeechobeensis ssp. okeechobeensis), and Rugel’s pawpaw (Deeringothamnus rugelii) that are listed as endangered in counties of the LSJRB (USFWS 2017a). Other threatened and endangered species are found in Section 4.4. Under review, the candidate Black Creek crayfish is found in Doctors Lake and Rice Creek (USFWS 2017a). Doctors Lake is listed as impaired due to nutrient loading and Rice Creek is impaired to due to dioxin levels from Georgia Pacific discharge (SRR 2016).

Urbanization and subsequent habitat loss and alterations can result in negative interactions between humans and wildlife. For example, the Wildlife Service is called in to disperse or dispatch a variety of animals. Between the years 2006 and 2011, gulls, egrets, and herons represented 57% of the 4,407,393 animals that the agency dispersed through a variety of measures in Florida (e.g., firearms, pyrotechnics, pneumatics, and electronics) (Levine and Knudson 2012). Cooper and Vanderhoff 2015 recorded greater numbers of the brown pelican at Mayport during autumn through spring months and along the river at Jacksonville University during winter and spring months, from a study conducted in September 2012 to August 2013. By comparison, numbers reported to eBird, a database monitored by the National Audubon Society and the Cornell Lab of Ornithology, were greatest during winter months. Comparing the Christmas Bird Counts in years 2000 and 2011 to 2016 from a marsh near Clapboard Creek, annual counts were generally greatest in 2011 (Table 4.6). 2016 counts were lower than 2015 counts for the bald eagle, snowy egret, wood stork, black skimmer, tricolored heron, little blue heron (Table 4.6; Audubon 2017). By comparison, the brown pelican and laughing gull increased by 100 individuals each. The white ibis increased by 300 individuals. Changes in counts may represent habitat modifications in nearby areas.

Table 4.6 Christmas bird counts of selected species from Jacksonville marsh site in 2000, 2011-2015. SSC – species of special concern; ST – state listed, threatened; FT- federal listed, threatened; FE – federal listed, endangered (Source: Audubon 2016).
Brown pelicanSSC634200012501100800600700
American oystercatcherSSC131478688
Laughing gull5123100950800110016001700
Bald eagle18343532214036
Piping ploverFT24846181213
Snowy egretSSC307391175175400450342
Wood storkFE120128105120100260140
Black skimmerSSC8100005403506001000649
Tricolored heronSSC128676050100175100
Little blue heronSSC54937580100200165
White ibisSSC35210004002009008001000
Roseate spoonbillSSC134561345

Least terns are migratory birds that require sandy or gravel habitats with little vegetation for nesting. Rooftop nesting sites have become more common due to habitat loss. Large rooftop populations have been recorded at NAS Jacksonville (Jackson 2013).  In Florida, Wildlife Service Agency had been called upon to disperse 273 least terns in 2011, indicating negative interactions with humans (Levine and Knudson 2012).

Wood storks (endangered) nest in the LSJR and feed on fish among other animals, requiring 450 lbs of fish per pair during the nesting season (SRR 2016). They require shallow pools that dry up to help concentrate fish prey. During extended periods of drought, wood stork numbers decrease. Currently, populations are considered stable and close to carrying capacity with respect to numbers of nesting pairs. Jacksonville Zoo and Gardens (91 nests in 2015) and Dee Dot (130 nests in 2016). Pumpkin Hill nests were active in 2009 but since then data have been unavailable or the site inactive, respectively (SRR 2016). Between the years 2000 and 2016, numbers of wood storks were greatest in 2015 and then decreased in 2016 in a marsh near Clapboard Creek (Table 4.6). In Florida, Wildlife Service has been called upon to disperse 270 wood storks in 2008-2011, indicating negative interactions with humans (Levine and Knudson 2012).

The following trends in wetlands within Florida and certain sections of the LSJRB are also notable:

  • In Florida, the conversion of wetlands for agriculture, followed by urbanization, has contributed to the greatest wetland losses (Dahl 2005).
  • The Upper Basin (the marshy headwaters of the St. Johns River) has experienced substantial historical wetland loss, and by 1983, it was estimated that only 65% of the original floodplain remained (SJRWMD 2000).
  • Hefner 1986 stated that “over a 50-year period in Northeast Florida, 62 percent of the 289,200 acres of wetlands in the Johns River floodplain were ditched, drained, and diked for pasture and crop production (Fernald and Patton 1984).”
  • According to DEP 2002, “the 1999 District Water Management Plan notes seven to 14 percent losses of wetlands in Duval County from 1984 to 1995, according to National Wetlands Inventory maps.”
  • In 2012-2013, the SJRWMD reported a loss of 380.7 wetland acres as compared to 14.5 acres created, 2,268.6 acres preserved, and 660.1 acres enhanced (DEP 2014f).

Duval Country is characterized with very high runoff values (57-331 mm, a ratio of urban runoff relative to county area) due to increases urbanization (Chen, et al. 2017). Current Status of Wetlands in Florida

A discussion of wetland status in the LSJRB is incomplete without an evaluation of wetlands within a broader, historical context. Although wetlands maps do not reveal with any statistical certainty how many acres of wetlands in the LSJRB have been gained or lost over time, there are reliable historical records in the literature that estimate how many wetland acres have been lost throughout the state of Florida over time. A literature search was conducted to compile comparable and quantifiable estimates of historical wetland change in Florida over time. Because data occurring within just the LSJRB could not be extracted from statewide data, information for the whole state of Florida was evaluated and compiled in Appendix 4.2.B.

Prior to 1907, there were over 20 million acres of wetlands in Florida, which comprised 54.2% of the state’s total surface area. By the mid-1950s, the total area of wetlands had declined to almost 15 million acres. The fastest rate of wetland destruction occurred between the 1950s and 1970s, as the total area of wetlands dropped down to 10.3 million acres. Since the mid-1970s, total wetland area in Florida appears to have risen slightly. Net increases in total statewide wetlands are attributed to increases in freshwater ponds, such as manmade ponds created for fishing, artificial water detention or retention, aesthetics, water management, and aquaculture (Dahl 2006). The average of all compiled wetlands data in Florida revealed that the state retained a total of 11,371,900 acres by the mid-1990s (occupying 30.3% of state’s surface area). This translates into a cumulative net loss of an estimated 8,940,607 acres of wetlands in Florida since the early 1900s (a loss of 44% of its original wetlands) (Appendix 4.2.B).

4.2.7. Current Trends in Wetlands in the LSJRB

Trends in wetlands can only be ascertained from sequential, time-series data. The only dataset of this type regarding wetlands within the LSJRB is contained within Land Use/Land Cover maps from the SJRWMD. These Land Use/Land Cover maps include spatial data on wetland types and were produced in 1973, 1990, 1995, 2000, 2004, and 2009. Trends in Total Wetlands Acreage

Acres per year of wetlands derived from the SJRWMD Land Use/Land Cover maps are not comparable or statistically robust in order to establish trends in total wetland acreage over time. The lack of comparability between years stems from differences in the techniques, scale, and wetlands interpretation. However, development pressures that result in wetland loss and function indicate a WORSENING trend in total wetland acreage within the LSJRB. Trends in Wetland Vegetation


Although the total wetland acreage cannot be statistically compared from year to year, the relative contribution of different wetland types can be statistically compared with an acceptable degree of reliability. These comparisons attempt to assess how the quality of wetlands in the LSJRB might have changed over time.

Most categories of wetlands used in the SJRWMD Land Use/Land Cover maps were not consistent over the years. Notably, the categories used in 1973 were markedly different from the categories used in the 1990-2004 maps (see Appendix 4.2.C.).

When wetland codes are grouped into two broad categories (forested wetlands and non-forested wetlands), significant trends are noted. There appears to have been a shift in the composition of wetland communities over time from forested to non-forested wetlands (Figure 4.5). Forested wetlands comprised 91% of the total wetlands in 1973, and constituted only 74% of total wetlands in 2009. Brown and Shi 2014 estimated freshwater forested wetlands represent twice the ecosystem value as non-forested wetlands (Figure 4.5). Because forested wetland categories changed over the years, specific categories that declined between 1973 and 2009 cannot be identified (Appendix 4.2.C.). However, between 2000 and 2009, almost every forested wetland category had decreased, with cabbage palm hammock and pond pine having the greatest decreases. By comparison, non-forested wetlands comprised 9% of the total wetlands in 1973 and 26% in 2009. Wet prairies and mixed scrub-shrub wetland categories contributed to observed increases (Appendix 4.2.C.). In the LSJRB between 2006-2013, forested wetlands represented 47-97% of permitted impacted wetland area per year (Goldberg and Reiss 2016).

Figure 4.5
Figure 4.5 Forested wetlands and non-forested wetlands in the Lower St. Johns River Basin based on land use/land cover maps (SJRWMD 2010b).

4.2.8. Wetland Permit Trends in the LSJRB

The SJRWMD process environmental resource permits that may impact wetlands and surface waters (SJRWMD 2014). In general, these projects were located in mixed hardwood wetlands. During 2016, 56 SJRWMD-processed permits that impacted wetland were issued and were each ≤101 impacted wetland acres. In St. Johns County, the Silverleaf Plantation development plans to build more than 2,000 homes on its property, impacting 314 wetland acres (SJRWMD 2016b). Between the years 2000 and 2014, the majority of issued permits was for <10 average impacted wetland acres in a project, based on SJRWMD permitting records (Figure 4.6). Those permits for residential and small business land use modifications highlight the challenges with wetland mitigation. Incremental wetland conversions result in cumulative impacts at the landscape level. In 2016, issued permits proposed to impact 753.7 wetland areas that were on average 12 acres each. Another 129 impacted wetlands areas did not require mitigation and were each 1.4 acre on average (SJRWMD 2016b).

Wetlands are fragmented across the urban landscape and different habitats occur within and surrounding project sites (Kelly 2001) which then impacts wetland function and community composition (Faulkner 2004). If wetlands are few and far between, then travelling amphibians and other animals are exposed to pollutants and death on roadways (Faulkner 2004).  Even smaller wetlands <0.2 ha contribute to local diversity (e.g., juvenile amphibians, Semlitsch and Bodie 1998).

Permits for modifying small wetlands are the largest in numbers and yet the contribution of these wetlands to local diversity and function remains undocumented (Figure 4.6; Semlitsch and Bodie 1998). Permits are given to individuals and are site specific, but cumulative impacts due to the number of conversions at the landscape scale are not addressed. At the landscape level, these smaller and isolated wetlands are not as valued as riverine wetlands (Brody, et al. 2008) and may not be protected by the Clean Water Act. Research is showing that these smaller wetlands can help take up nutrients via denitrification processes and thus reduce nitrogen and phosphorous, particularly in areas where there is heavy nutrient loading (e.g., agricultural and urban locations) (Lane, et al. 2015). In addition, smaller wetlands contribute to the buffering of the local water table, in part due to the cumulative exchange along the perimeter of many smaller wetlands as compared to fewer but larger wetlands (McLaughlin, et al. 2015).

Figure 4.6
Figure 4.6 Numbers of SJRWMD permits per project impacted wetland acreage from 2000 to 2016 (SJRWMD 2016b).

Based on SJRWMD permit records, the methods used to mitigate wetlands have changed over time (Appendix 4.2.D.). During the early 1990s, wetland areas were most commonly mitigated by the creation of new wetlands or through wetland restoration. During the 2000s, relatively few wetlands were created or restored with most mitigation occurring through the preservation of uplands/wetlands (Figure 4.7). In 2016, permitees of 66 projects applied for a total of 179.5 credits, and 11 projects were permitted for on-site only mitigation (SJRWMD 2016b). No permitees planned for enhancement or restoration.

Figure 4.7
Figure 4.7 Percentage (bars) of issued permits that opted for purchasing mitigation credits, wetland preservation, creation, upland preservation, or enhancement and total impacted wetland acres (line) in the years 2006, 2013 to 2016, indicating in parentheses the total number of permits issued for mitigated impacted wetlands. Because permittees may opt to use more than one type of mitigation for a project, total percentages per year will exceed 100% (SJRWMD 2016b). Trends in Wetland Acreage Impacted/Mitigated by Permits Granted by USACE

For a complete analysis of wetlands impacted and mitigation in the LSJRB, data needed from the USACE would include the location, total acres, type of vegetation, maturation/stage of wetland, wetland functions replaced, and wetland services replaced. A similar data deficit was found by the NRC, which concluded that “data available from the Corps were not adequate for determining the status of the required compensation wetlands” (NRC 2001).

In 2016, the trend continued for mitigation banks to be proposed for mitigation of wetland dredge and fill activities rather than for preservation, creation, enhancement, or restoration (Figure 4.7). The mean ratio of mitigation credit per acreage of impacted wetland was greater in 2006 (2.1, n = 29 permits) than in 2015 (0.48, n = 38 permits) and 2016 (0.49, n = 41 permits) for those projects that used mitigation banks as the only type of wetland mitigation. The annual percentage of permits issued that proposed to purchase mitigation credits was lowest in 2006, was approximately double the percentage in 2015, and was also relatively high in 2016 (Figure 4.7).

The percentage of permitted projects that planned for wetland preservation was greatest in 2014 as compared to 2006 and 2016 (Figure 4.7). The largest project Silverleaf Plantation was permitted to impact 314 wetland acres, preserve 1,931 wetland acres and 366.6 upland acres, and purchased 28.9 mitigation credits (SJRWMD 2016b).

Table 4.7 Acreage comparison in wetland mitigation for permits issued in 2006, 2013, 2015, and 2016 that required mitigation (SJRWMD 2016b).
110 Permits
47 Permits
69 Permits
66 Permits
Impacted wetlands774434274754
Wetland preservation48536915382384
Upland preservation100113984429

With the modification of Mile Point for navigational safety, USACE proposed to use the dredge spoils for restoring 53 acres of wetland habitat to Great Marsh Island. In April, 2016, dredging of the waterway was completed (Werely 2016). Loss of 8.15 acres of salt marsh at Helen Cooper Floyd Park would be mitigated at Great Marsh Island that has experienced erosion. (USACE 2011). In addition, up to 26.2 acres of salt marsh would be restored. The salt marsh species would be planted across 8 (costing $465,888) to 53 (costing $3,029,701) acres, depending on the plan chosen. Another 8 acres of restored marsh would be from dredging of the proposed Flow Improvement Channel in Chicopit Bay. In addition, deploying the spoils at Great Marsh Island would save $9,056,000 as compared to disposing at Buck Island (USACE 2011).

The Supreme Court ruled on the Koontz v. St. Johns River Water Management District case (USSC 2013). Koontz, a Florida landowner had challenged the terms set by the SJRWMD for filling in wetlands on his property near Orlando (Sommer 2013). The SJRWMD had proposed a plan for mitigation on district lands rather than on-site as proposed by Koontz. Koontz withdrew his application, the permit was denied, and the lawsuit was filed. The Supreme Court ruled in favor of Koontz. There is concern that regulators may be less likely to negotiate with land developers to avoid possible lawsuits and instead, may deny permits (Sommer 2013).

4.2.9. Future Outlook


HIGH VULNERABILITY. The total spatial extent of wetlands negatively impacted through the SJRWMD permit process is increasing each fiscal year and is likely to increase with the improvement in the national and state economies. These impacts are magnified by the losses of wetlands permitted by the USACE (the evaluation of these Section 404 permits is limited in this study). Many remaining wetlands are susceptible to alteration and fragmentation due to growing population pressures in northeast Florida. Urbanization at the landscape level has a direct impact on wetland communities. For example, between 2006-2013, approximately 73% of the 1,046 ha of impacted wetlands were located in Mid to High Development and 18% in Mid Development parcels (Goldberg and Reiss 2016). Incremental filling of depressional ponds in addition to developing along waterways have the consequence of altering local hydrology, adding nutrients and heavy metals to the sediments and water column, bioaccumulation of heavy metals up the food web, and increasing the number and coverage of nuisance and invasive species. Isolated wetlands can retain 1,619 m3 water/ha, on average, from models developed for Alachua County, FL, wetlands (Lane and D’Amico 2010). The potential for flooding, hydrologic alterations, and pressures on species diversity will continue with the loss of wetlands in the LSJRB.

In addition to development and withdrawals, tidal wetlands will be impacted by sea level rise. Tidal wetlands in the river are unlikely to outpace sea level rise estimated at 3 mm/year (Weston 2014) due to inability of marsh vegetation to accrete organic material at faster rates. Delivery of fluvial suspended sediments is relatively low in the St Johns River, compared to other U.S. rivers (Weston 2014). Turbidity in the mainstem is improving, indicating that sediment export to the tidal wetlands is low (see Turbidity section; SRR 2016). In 2015, the maximum turbidity value in 2015 was the lowest since 1997 (SRR 2016). Coastal wetlands may be less impacted by sea level rise. Contrary to expectations of coastal erosion with sea level rise and disruption of longshore drift with dredging activities, shorelines along Duval and St Johns counties have been advancing since the 1800s (Houston and Dean 2014). On-shore sediment deposition is the likely mechanism and may help buffer erosion and sediment transport due to sea level rise in the future (Houston and Dean 2014).

Wetlands in the LSJRB will be affected in the future due to surface water withdrawals from the river as permitted by the SJRWMD. In order to fully understand and predict the potential effects, the SJRWMD released the St. Johns River Water Supply Impact Study in February 2012 after a peer review by the National Academy of Sciences — National Resource Council (SJRWMD 2012a). In this study, the St. Johns River was divided into segments for analysis – the first three of which fall into the LSJRB:

SEGMENT 1 (“Mill Cove”) – extends 39.6 km from Mayport to the Fuller Warren Bridge.
SEGMENT 2 (“Doctor’s Lake”) – extends 25.4 km from the Fuller Warren Bridge south to a line close to Fleming Island.
SEGMENT 3 (“Deep Creek”) – extends 98.1 km from Fleming Island to Little Lake George.

The expected impacts to wetlands in the above segments of the LSJR were analyzed under four different modeling scenarios. One scenario was constructed to create a baseline that was used directly to assess salinity changes. Three scenarios were based on modeled data, a full water withdrawal, and various treatments of land use data, Upper SJRB projects, and sea level rise (SJRWMD 2012a). According to the SJRWMD (SJRWMD 2012a), the overall results were that “some specific wetland types were reduced in area under each scenario. However, loss in total wetland area was not shown under any scenario with any of the analytical approaches used” (SJRWMD 2012a, p 10-80). More specific results of the study are summarized below.

Based on the modeling results, each segment within the LSJRB is expected to experience a change in annual mean salinity, which would, in turn, affect wetland communities. River Segment 1 is predicted to experience a change in mean annual salinity of 0.32 psu, followed by a 0.12 psu change in Segment 2, and 0.011 psu change in Segment 3. The likelihood of salinity effects in Segments 1 and 3 were deemed to be “low,” because Segment 1 is already dominated by saltmarsh species, which would tolerate the increase in salinity without negative impacts. The increase in salinity in Segment 3 was very small and was not expected to cause noticeable shifts in vegetation. However, river Segment 2 is considered the area of greatest concern, because this area between the Fuller Warren Bridge and the Shands Bridge is dominated by hardwood swamps and extensive areas of freshwater and transitional vegetation. In this segment, salinity effects were deemed to be “high.”

The St. Johns River Water Supply Impact Study also evaluated changes in patterns of water inundation and water depth (SJRWMD 2012a). However, the segments contained within the LSJRB were not analyzed for change in stage, because water levels in the LSJR are so heavily influenced by sea level. According to this study, the modeled water level change in the Segments 1-4 due to water withdrawals was less than 1 cm. Throughout the entire SJR, the average depth change ranged between 4 cm to less than 2 cm depending on the scenario used. The category of wetlands most negatively impacted throughout the state was “freshwater marshes.”

Using the Ortega River as a model system, the St. Johns River Water Supply Impact Study examined whether surface water withdrawals could potentially cause movement in the freshwater/saltwater interfaces along the river. SJRWMD researchers identified sampling stations along the Ortega River and conducted vegetation studies. They determined five main wetland plant communities along a gradient from freshwater to brackish water: Hardwood Swamp, Tidal Hardwood Swamp, Lower Tidal Hardwood Swamp, Intermediate Marsh, and Sand Cordgrass Marsh. The soil salinity breakpoints and river salinity breakpoints, where one plant community type shifts to another type, were determined (Table 4.8).

Table 4.8 Soil and river salinity breakpoints causing wetland vegetation shifts in the St. Johns River Basin, Florida (as determined in SJRWMD 2012b).
Soil Salinity
River Salinity
Predicted Distance Moved in St. Johns River
0.47 psu3.22 psu2.83 km
1.53 psu4.13 psu3.10 km
2.44 psu4.93 psu3.30 km
3.41 psu5.77 psu3.34 km

The study predicted upstream movement of vegetation boundaries of up to 1.13 km along the Ortega River. When the Ortega River model was applied to the entire St. Johns River, the directional shift of wetland vegetation community types ranged from 3.34 km to less than 0.21 km (SJRWMD 2012a).

Thus, certain types of wetland communities will be negatively impacted by future surface water withdrawals in the St. Johns River. These impacts must be considered cumulatively with other expected impacts from future changes in land use, surface water runoff, rainfall, navigational works, groundwater, and sea level rise.

QUESTIONABLE QUALITY. Further investigation is needed to determine the quality and longevity of mitigated wetlands and their ability to actually perform the ecosystem functions of the wetlands they “replace.” An increasing proportion of these mitigation wetlands represent uplands/wetlands preserved on average >30 miles from project site (Brody, et al. 2008), including many acres in wetland mitigation banks. If preserved wetlands represent already functional wetlands, then they do not replace the ecosystem services lost to development. Currently, there is no accounting of the specific locations of each impacted wetland.

Restored and created wetlands generally do not reach ecosystem functioning present in reference wetlands. Based on a meta-analysis from published studies of 621 wetlands, Moreno-Mateos et al. (Moreno-Mateos, et al. 2012) reported that ecosystem services were not returned with restoration efforts in either created or restored wetlands. The size of the wetland (>100 ha) recovered more quickly than smaller wetlands (0.1, 1, and 10 ha). Wetlands only reached on average 74% of biogeochemical functioning after 100 years. In addition, plants and vertebrate diversities in restored/created wetlands remained lower than reference wetlands after 100 years. By comparison, macroinvertebrates reached references assemblages between 5 and 10 years. In comparing different types of wetlands, riverine and tidal wetlands recovered more quickly (up to 30 years) as compared to depressional wetlands that did not reach reference conditions (Moreno-Mateos, et al. 2012).

Wetlands at the mitigation banks are not necessarily reaching a measure of success relative to reference conditions. Difficulties in restoring wetlands may be related to past activities on the property and indirect effects due to surrounding land use. For example, land use at Loblolly, Tupelo, and Sundew mitigation banks were previously agricultural, managed pasturelands, and mixed agriculture and/or low intensity urban, respectively (Reiss, et al. 2014). Reiss, et al. 2007 investigated success and compliance of 29 wetland mitigation banks in Florida. Barberville, Loblolly, Sundew, and Tupelo were included in their study (Tables 4.3 and 4.4). These mitigation banks did not include a target for success criteria or a reference condition (either a reference database and/or comparison sites, Reiss, et al. 2009) to measure success (e.g., wildlife needs). With respect to exotic and nuisance cover, final success criteria for state permit requires <10% exotic and nuisance cover (except for Barberville: 5% exotic, 10% nuisance). Reiss, et al. 2007 recommend that monitoring should also encompass flora and fauna, and not just exotic and nuisance species. At the time of their study, Barberville was a ‘long ways off’ from final success due to pines having to be replanted. Loblolly and Tupelo had started plantings and was described as not communicating so well in providing the monitoring and management status reports. Sundew was also described as not communicating so well with reports (Reiss, et al. 2007). Reiss, et al. 2007 argue that functional equivalency in wetland mitigation banking remains questionable without a clear method to assess ecosystem function. LDI scores within the mitigation banks indicate that wetland function may be impossible to achieve (Reiss, et al. 2014).

The USACE and the EPA have released new rules regarding compensatory mitigation of wetlands impacted by USACE permits (took effect on June 9, 2008). According to the Federal Register, the new rule emphasizes “a watershed approach” and requires “measurable, enforceable ecological performance standards and regular monitoring for all types of compensation” (USACE 2007). How these new changes may or may not affect wetland mitigation in the LSJRB warrants future investigation. Given the connectivity of aquifer and ground water via fracture lines, those activities that uptake water in one location may prevent the watershed from being recharged during precipitation events and exacerbate drought effects on wetland systems (Bernardes, et al. 2014).

Partial restoration of riparian corridors can have fairly immediate and positive impacts on nutrient levels and diversity of local flora and fauna (Rossi, et al. 2010). The authors had planted riparian species of trees, shrubs, grass, and forbs to increase structural complexity in areas 3 x 9.5 m along first-order tributaries of the LSJR. After three months, sampling was conducted for two years. Macroinvertebrate diversity increased (Coleoptera and Lepidoptera), dominance of pollution-tolerant taxa decreased, and pollution-intolerant taxa (Odonta and Ephemeroptera) increased as compared to non-restored sites. In addition, soil nitrate was significantly less in the restoration sites than control sites and soil phosphorous decreased over time in restored sites due to nutrient uptake by the plants. The authors recommend incorporating restoration areas along urban stretches of the river to promote ecosystem function (Rossi, et al. 2010).

The Lasalle Bioswale Project showcases another way to minimize contaminants from entering waterways. Bioswales are vegetated areas that collect stormwater runoff. Plants and soil communities take up the pollutants and thereby treat pollutants found in stormwater runoff. This particular project was accomplished by the St. Johns Riverkeeper and partners (St. Johns Riverkeeper 2013b).

In summary, the future outlook for the health of the LSJRB depends upon detailed, accurate, consolidated record-keeping of wetland impacts, the cumulative impact of parcel-by-parcel loss of wetland ecosystem functions and services, and the success of wetlands enhanced, created, or restored. Given the continued trend of mitigation via purchase of mitigation credits and off-site conservation areas in place of on-site mitigation, the outlook for local wetlands in the LSJRB does not look promising.