2.0 Report of the Soils Subgroup · 2013-03-18 · Although always underfoot, soil is often a less...

The SERES Project Soils Subgroup Report

74

2.0 Report of the Soils Subgroup T.Z. Osborne, S.E. Davis, G.M. Naja, R.G. Rivero, and M.S. Ross

Problem Overview _____________________________________________________________________________

Central Questions: How do hydrology and water flow affect the carbon balance, peat accretion

and/or subsidence in Everglades soils? Further, how do other soil nutrients (mainly N and P) and

their abundance affect Everglades soil stability?

Given the potential magnitude of sea-level rise over the next century and the existing issues with

Everglades water quality and quantity, answers to these basic questions are needed in order to

effectively manage the soil resources of this expansive south Florida ecosystem. Understanding

the controls on the dynamics of the Everglades’ organically derived soil platform will allow us to

more effectively prevent soil loss and may even provide us the capacity to “grow” parts of the

ecosystem in concert with sea-level rise. Although answers to these questions can be most effec-

tively addressed through controlled experimentation and continued field observation, a thorough

review and synthesis of existing science on Everglades soils can help to serve as the basis for the

development of research hypotheses and experimental treatments.

Although always underfoot, soil is often a less recognized or underappreciated ecosystem com-

ponent due, in part, to the subtle nature of change associated with it. Changes in wetland soil

characteristics, whether in nutrient content, metals, organic contaminants, or simple changes in

physical properties, are not readily apparent to even the trained observer. Rather, changes in

other ecosystem components, such as vegetation or water chemistry, that can be modulated by

soil quality are often the first cues that ecosystem changes in soils have occurred.

Soils, by nature, are long-term integrators of environmental condition. In the Everglades ecosys-

tem, soil also serves as a major storage pool for organic matter, major nutrients (C, N, and P),

metals (both beneficial and harmful), and anthropogenically derived organic substances. The


75

physical position of soil can influence landscape patterns (ridge, slough, tree islands) and it also

serves as a medium of growth for most vegetation in the system. Further, the interactions be-

tween water level (natural or managed) and soil elevation (affected by accretion or oxidation) at

a given location in the landscape are of great importance in controlling nutrient and plant com-

munity dynamics.

Because of the significant role soils play in the health and function of the Everglades ecosystem,

soil conservation is an important aspect of Everglades Restoration. As such, it is critical to antic-

ipate ecosystem responses to hydrologic restoration activities from the perspective of the soil

component. The broad goal of this work is to synthesize the best available scientific understand-

ing of the drivers of soil condition in the Everglades ecosystem and thus anticipate the resulting

effects of various hydrologic restoration scenarios. To communicate this effectively, a general

review of soil related science in the Everglades is warranted.

This section of manuscript reviews pertinent scientific literature relating to Everglades soils and

thus serves as an introduction and foundation for discussion of anticipated reactions of the soil

component of the ecosystem to hydrologic restoration in the future. The following discussion

will cover the types of soils found in the Everglades landscape, the role of soils in ecosystem

function, the factors governing soil formation and maintenance, human impacts to Everglades

soils, and management implications of restoration with respect to soil.

Soil types in the Freshwater Everglades landscape _____________________________________________________________________________

There are three major soil types found in the Everglades: 1) peat soils, which are high in organic

content and are comprised of partially decayed plant material, 2) marl soils, which have lower

organic content and are comprised of calcitic mud deposited from calcareous periphyton

(Gleason and Spackman 1974) and 3), tree island soils that have greater portions of mineral

components than peat soils, but are very similar in that their origins are both plant material. The

origin and development of peat and marl soils are greatly dependent upon water depth and result-

ing wetland vegetative communities. Alteration of the hydroperiod and diminished surface water


76

inundation may also alter the vegetation communities and subsequent changes in soil type and

depth or elevation could occur.

The earliest classification of soil in the Everglades was based on the vegetation encountered in

the strip of land between Lake Okeechobee and the eastern edge of the Everglades. In 1915,

Baldwin and Hawker (1915) applied the term “sawgrass peat” to the soil underneath the land

with a wide expanse of sawgrass. Davis and Benett (1927) defined peat as a material containing

more than 65 % plant remains with very little mineral matter. The soils of the Everglades were

extensively surveyed (fieldwork) in the 1940s by John (1948) differentiating the peat types based

on the nature of the organic materials and the kind and depth of underlying materials.

Peat Soils

Peat soils in the Everglades occur in areas where the bedrock is deeply recessed from the surface.

Its formation requires anaerobic conditions brought about by long hydroperiods and it is com-

posed predominantly of organic remains of dead plants. Two types of peat can be found in the

Everglades: Everglades and Loxahatchee peat types encompassing over 7000 km2. Table 2-1

summarizes some chemical and physical characteristics of these two soil types (also known as

histosols). Everglades peat underlies much of the central and southern Everglades and Loxa-

hatchee peat occurs in deeper marsh areas underlying LNWR, the northeastern areas of WCA-2A

and the Shark River Slough (Gleason and Stone 1994). The Everglades peat is formed mostly

from partial decomposition of sawgrass, the dominant wetland plant species whereas Loxa-

hatchee peat forms from vegetation of sloughs, especially water lily. Everglades peat is a brown

to reddish brown fibrous peat (Snyder 1994), with the darkness attributed to charcoal matter

from frequent fires (Lodge 2010). Loxahatchee peat consists of loose or spongy brown fibrous

peat. It is lighter colored than Everglades peat and contains the remains of slough vegetation.

The peat thickness is around 3.0 m in LNWR and decreases to a minimum of 0.4 m in the ENP

(Craft and Richardson 2008). Peat soils are subject to subsidence and loss of surface elevation

when drained. Oxidation and compaction are considered the dominant subsidence forces, and

from a practical standpoint are irreversible (Scheidt and Kalla 2007). The R-EMAP Program

previously documented soil subsidence in the public Everglades. It has been established that


77

from the 1940s to 1990s the entire Everglades Protection Area lost up to 28 % of its soil volume

due to soil oxidation and subsidence (Scheidt and Kalla 2007).

Craft and Richardson (2008) performed a carbon-14 dating of peat collected from similar depths

in four cores from LNWR, WCA-2, WCA-3 and ENP. It revealed a gradient of increasing peat

age of the peat along the north-to-south gradient. Subsurface peat in the LNWR and WCA-2A

was deposited only 500 and 800 years BP, respectively. In WCA-3A and ENP, peat collected

from the same depth was deposited more than 2,000 and 2,500 years BP, respectively. The large

difference in the 14C age of peat collected from similar depths from the LNWR, WCA-3A, and

ENP suggests that historical environmental factors or rates of accretion differed greatly between

the northern and southern Everglades. The bedrock underlying the southern Everglades consists

of Miami limestone that is more permeable than the Fort Thompson formation that underlies are-

as of the northern Everglades (Gleason and Stone 1994). The porous nature of the Miami lime-

stone perhaps contributes to reduced hydroperiod and, consequently, increased fire frequency in

the southern Everglades, leading to reduced peat accretion and thinner peat at these locations

(Craft and Richardson 2008).

Table 2-1. Characteristics of Everglades peat soil and Loxahatchee peat soil (Craft and Richardson 2008).

USDA

classification

Area

(km2)

Thickness

(m)

Origin pH Organic

content

(%)

Everglades

peat

Euic hyperthermic

typic medihemist

(Pahokee series)

4,420 0.5–2.0 Sawgrass (Cladi-

um jamaicense)

5.5–6.5 85–92

Loxahatchee

peat

Euic hyperthermic

typic medisaprist

(Terra Ceia series)

2,950 2.0–3.0 Water lily

(Nymphaea odora-

ta)

5.0–6.5 92+


78

Marl Soils

Marl soil is commonly found in the shallower peripheral marshes of the Everglades subject to

shorter periods of surface water inundation and is the main soil of the wet prairies near the edges

of the southern Everglades. The marl soil is commonly found in association with periphyton (al-

gal assemblage), which precipitates calcium carbonate from the water column as a consequence

of photosynthesis and high pH (Scheidt and Kalla 2007).

The southern marl prairies, which are predominantly composed of shallow marl soils situated

above limestone, are a mosaic of wet prairie, sawgrass, tree islands, and tropical hammock com-

munities that support a high diversity of plant species (Olmsted and Loope 1984). Olmsted and

Armentano (1997) summarized the various terms used by a number of earlier authors to describe

the wet prairies found in the Shark Slough region in ENP. Kushlan (1990) stated that the “wet

prairie is the least frequently flooded of any Florida marsh type. Their short hydroperiod (50 to

150 days per year) preclude peat development.” Finally, Olmsted and Armentano (1997) report-

ed that wet prairies refer to marl areas dominated by mixtures of forb and graminoid species but

with the dominance of sparse sawgrass and spikerush and increasing admixtures of many other

species associated with slightly higher elevations and sometimes rock outcrops.

Included in this landscape are the Rocky Glades, which are north of Long Pine Key. These areas

support the driest vegetation (Kushlan 1990) and typically have a high frequency of limestone

exposures above the marl soil, resulting in what is locally called a micro-karst topography

(USFWS 1999). In some areas of the rocky glades region and in the peripheral wet prairies to

Taylor and Shark sloughs, there are numerous solution holes in bare limestone retaining organic

matter, and over long periods of time, develop a soil locally known as "Gandy peat" within them

(USFWS 1999).

Tree Islands Soils

Tree islands are landscape features of the northern Everglades areas and Shark River Slough,

formed on areas of slight topographic relief. Depending on their location, tree islands may be of

three general soil types: a) frequently flooded peats, characteristic of tree islands supporting


79

swamp forest species throughout the Everglades, including ghost tree islands of the WCAs (Ewe

et al. 2010); b) shallow, organic, relatively low-P soils, common on rarely flooded sites in the

seasonally-flooded marl prairie landscape, and supporting tropical hardwoods; and c) deeper,

alkaline, mineral soils with extremely high P concentrations, most common in infrequently

flooded hardwood hammocks embedded in long hydroperiod marshes of ENP and WCA-3.

Soil phosphorus content of tree islands range widely (500 - 10,000 ug.g-1; Wetzel et al. 2009,

Ross and Sah in review), but are usually considerably higher than soils in the surrounding land-

scape (Wetzel et al. 2005; Richardson et al. 2008). There are various hypotheses on how tree

islands form and maintain based on the wide range of nutrient inputs (precipitation, plant litter,

bird guano deposition, groundwater upwelling, and transpiration of tree island hammocks (Wet-

zel et al. 2005; Givnish et al. 2008).

Role of soils in ecosystem function ______________________________________________________________________________

Soils are an important, although often underappreciated, component of almost any ecosystem and

as such, play both a structural and functional role in ecosystem dynamics. This is especially the

case in the Everglades where the unique environmental conditions and biogeochemical proper-

ties of the soils are integral to maintenance and function of the ecosystem.

Structural Role of Soils

At the most intuitive level, soil is the three-dimensional natural body that forms the substrate or

surface material in which vegetation is rooted and upon which both flora and fauna are estab-

lished. This medium, or substrate, may span the spectrum from mineral (by mass, the majority

of material in soil is mineral) to organic (very large portion of soil is organic matter) in nature

(Brady and Weil 2003). As mentioned previously, the organic soils of the Everglades, such as

the Loxahatchee and Everglades peats, are highly organic in nature with loss on ignition (LOI)

values ranging from 50-97% (Reddy et al. 2005; Scheidt and Kalla 2007).


80

Because of the predominantly anaerobic conditions brought about by hydrology, most soil for-

mation is organic in nature. Substantial plant productivity, coupled with slow anaerobic decom-

position, results in accretion of organic matter in the soil pool (Figure 2-1; Reddy and DeLaune

2008). In areas where hydroperiod is truncated to less than approximately 5 months a year, and

thus accretion of organic matter from plants and periphyton is very low, marl soils dominate

(Gleason 1972; Chen et al. 2000). Calcareous periphyton tissue is rapidly decomposed under

aerobic conditions leaving the dominant soil material inorganic (calcium carbonate) with some

residual organic matter. Marl soil is significantly more stable with respect to change over time,

as the mechanism of removal is erosion or dissolution in contrast to the mechanism for peat re-

moval, which is oxidation.

While rainfall is plentiful in the Everglades, very low slopes and paucity of adjacent easily erod-

ible material results in little mineral soil material inputs to Everglades marshes. Hence, the land-

scape within and adjacent to the Everglades is fairly stable given normal hydrologic conditions.

The absence of topographic relief implies rare erosion-producing flow velocities. However,

there is evidence that material transport within Everglades sloughs may help maintain the corru-

gated landscape in the ridge and slough mosaic (Larsen and Harvey 2010a). DeAngelis and

White (1994) suggest that erosional processes brought about by sea level rise may be significant

in coastal soils (wave action) and climatic changes may contribute to erosion of tree island soils

from intensive rainfall (compressed rainy season).

The modern day Everglades sits atop a unique geologic framework, the result of thousands of

years of shell and coral deposition in a shallow tropical sea. Similarly, present day soils were

formed over thousands of years of organic matter and marl deposition in a shallow tropical

freshwater wetland basin (Petuch and Roberts 2007). As discussed previously, peat or marl for-

mation is a direct result of hydrology, which in an undisturbed Everglades system, was a direct

result of landscape position. For example, in ENP, peat soils exist in the Shark Slough drainage

feature due to lower bedrock elevation and resultant longer hydroperiods, whereas to the east and

west of Shark Slough, higher elevation bedrock results in shorter hydroperiods and thus marl

soils dominate (Lodge 2010; Osborne et al. 2011b). Peat and marl soils reflect very different hy-

droperiods in close proximity, however, as a general rule, significant ecosystem scale elevational


81

gradients in the Everglades system are rare (McVoy et al. 2010; Holt et al. 2006). The exception

to this rule occurs in the local scale transition areas from tree islands to marsh or from sawgrass

ridge to open water slough.

Soil elevation at the local level is a function of geology, hydrology, and biology. As mentioned

previously, local geologic foundations dictate, to a large degree, the hydrology or hydroperiod of

a given marsh area. This hydroperiod in turn directly influences the rate of soil accretion, which

is dictated by the opposing biological processes of primary production and respiration (Figure 2-

2a; Cohen and Spackman 1984). The quality (bioavailability) of organic matter which makes up

the soil can also determine rates of accretion as different plant types contribute organic matter of

variable quality to the soil (Osborne et al. 2007; Cohen et al. unpubl. data). During the last 1200

yrs, the accretion rate of peat in the northern Everglades has been found to be approximately 1.6

mm yr-1 (Gleason and Stone 1994; DeAngelis 1994) under inundated (anaerobic) conditions.

Under aerobic conditions, rapid decomposition can cause soil loss (i.e., negative accretion) of

approximately 3 cm yr-1, highlighting the significantly asymmetrical rates of the two competing

processes of production and decomposition (Stephens and Johnson 1951; Snyder and Davidson

1994; Maltby and Dugan 1994) and the integral role of hydrology in soil dynamics.


82

The key role of hydrology is that of modulator for soil elevation as low water (soil exposure to

aerobic conditions) can bring about loss of soil elevation due to primary (loss of buoyancy) and

secondary (microbial oxidation) subsidence or fire (abiotic oxidation). To date, severe soil oxi-

dation, and thus loss of elevation, has been noted in northern WCA-3A and WCA-2A and 2B

(Scheidt et al. 2000; Scheidt and Kalla 2007). By calculating the distance between peat eleva-

tions from the pre-drainage surface, using the Natural System Regional Simulation Model

(NSRSM) and the current condition (USACOE), mean subsidence of 1.7 m has been estimated

across the EAA and between 0.01 m and 0.9 m in the rest of the system, as shown in Figure 2-2b

(Aich and Dreschel 2011). This is on par with the range of 1.2 m to 1.5 m in the EAA presented

in Snyder (2005) and extreme localized subsidence > 3 m in some areas under sugarcane and

vegetable production (Snyder and Davidson 1994). These observations highlight the devastating

effects of prolonged dewatering on organic soil elevation in the Everglades.

Figure 2-2a. Schematic depicting deposition or peat building material. As annual deposition continues, previous years material is further decomposed and compressed under new material. Water levels modulate this process by reducing decomposition (flooded) or accelerating it (drained). Artwork by P. Inglett from Reddy and DeLaune (2008).


83

Figure 2-2b. Visualization of peat surfaces contours reflecting peat loss. Pre-drainage peat surface (from

Natural Systems Regional Simulation Model, NSRSM) is shown in grey-scaled colors. Cur-rent peat surface (from USACOE) is shown in the color-scaled below. Figure from: Aich and Dreschel (2011).

Functional Role of Soils

Soil, by way of its physical and biogeochemical properties can regulate two key ecosystem prop-

erties in the Everglades, plant community structure and nutrient cycling within the ecosystem.

Hydrology comes into play significantly with respect to both. In the case of vegetation communi-

ties, it is well known that each species of wetland plant in the Everglades has a range of tolerance

for flooding (Lodge 2010). This flood tolerance is a function of both depth and duration, either

of which can be detrimental to the species found in the system if extreme in either direction

(Mitch and Gosselink 2000).

In the Everglades, patterns of vegetation are often a result of elevation that in turn influences hy-

drology, creating a significant feedback loop. For instance, marl prairies are higher elevation,

and thus have shorter hydroperiods, and a distinct vegetation community structure that is suited

to very shallow soils with seasonal inundation (Osborne et al. 2011b; Davis et al. 2005; Browder


84

and Ogden 1999; Obeysekera et al. 1999; Olmstead and Loope 1984). Similarly, in areas with

longer hydroperiods, such as the ridge and slough systems in the central Everglades, soil eleva-

tion is very distinct in its effect on vegetation patterns. Soil elevation of sawgrass ridges, domi-

nated by Cladium jamaicense Crantz, is only slightly higher than that of the surrounding open

water sloughs (10-20 cm; Watts et al. 2010; Lodge 2010). This elevation difference delineates

the two competing landscape ecotypes and is reinforced (via feedback mechanism) by the quality

of the organic matter derived from the dominant vegetation types within ridges and sloughs

(Clark and Reddy 2003; DeBusk and Reddy 2003). Detrital material derived from sawgrass has

a significantly higher portion of the recalcitrant biopolymer lignin as opposed to the detritus de-

rived from Nymphea odorata (dominant deep water species) and associated submerged aquatic

vegetation commonly found in sloughs (Osborne et al. 2007; Clark and Reddy 2003). Thus po-

tential soil accretion rates in sawgrass ridges are nearly an order of magnitude higher than in

sloughs because of vegetation-derived differences in organic matter production rates and litter

lignin content. Seasonal hydrology serves as a feedback mechanism to reduce elevation on

ridges when drawdown allows aerobic conditions to accelerate enzymatic lignin degradation

(Criquet et al. 2000; Freeman et al. 2001). The hydrologically mediated interplay of soil accre-

tion and oxidation also serves to regulate ecosystem nutrient availability via organic matter stabi-

lization and mineralization, respectively (Figure 3-3).

In organic matter accreting systems such as the Everglades, the soil can serve as a sink or storage

pool for ecologically significant nutrients and elements (DeBusk et al. 1994; Newman et al.

1997; Bruland et al. 2006; Scheidt and Kalla 2007). Soils can also serve as a source in the bio-

geochemical cycling of these nutrients, and thus their ecological role as biogeochemical modula-

tors is significant (Reddy et al. 2005; Reddy and DeLaune 2008). As plants grow and access es-

sential nutrients from the water column or the soil, they are stored in the organic tissues of the

plant. After senescence, plant litter, the most significant contributor to soil accretion, still con-

tains relatively large amounts of these nutrients. The cycle from soil to plant to soil may be re-

peated indefinitely if not for the potential for some portion of these nutrients to be exported to

the surrounding water column. This export process can occur several ways, however, minerali-

zation of organic matter is usually a key step. The nutrients bound up in the soil organic matter

can be released to the water column when soil is oxidized by microbial activity or fire. Similar-


85

ly, the reverse is also possible. Elevated levels of nutrients or other constituents can be incorpo-

rated into soil organic matter via microbial uptake and plant growth. In some cases, simple con-

centration gradient driven mass transfer may enable soil enrichment of nutrients or export to the

water column.

Figure 2-3. Conceptual model depicting the interactions of plant community, nutrient loading, and hy drology with organic matter production. Biogeochemical cycles of major nutrients such as nitrogen, phosphorus and sulfur all revolve around the cycling of organic carbon. Artwork by P. Inglett from Reddy and DeLaune (2008).

Factors governing formation and maintenance of soil ______________________________________________________________________________

South Florida wetland soils of today reflect a developmental history dating back about 6 thou-

sand years, to a time when rapidly rising seas and a moistening climate combined to initiate the

formation of the Everglades on a pitted limestone surface. The range of soils we see today can


86

be explained by spatial variation in the five classical soil-forming factors, i.e., parent material,

geomorphology, climate, and the biota, operating over time (Jenny 1941). However, in wetlands

subject to hydrologic management, factors such as water flow, stage, duration, and quality must

also be accounted for collectively as a distinct soil-forming factor. Finally, wetland soils can be

sensitive to certain disturbances, e.g., major floods, fires, or hurricanes that physically move,

consume, or reshape them (Keddy 2000).

Though all of the above factors play a role, most Everglades soils are the products of biologically

mediated, hydrologically driven processes occurring within the basin itself over recent geologic

time. Figure 2-4 illustrates the major processes affecting soil development within this highly

regulated wetland. In the model, hydrology influences soil development in two ways: (1)

through in situ, biological processes that affect the balance between organic matter or calcite

production, which builds the soils up, and peat oxidation, which break them down; and (2) by

redistribution of materials from one place to another in the marsh, i.e., erosion and deposition.

The climate of a given year affects the amount of water in the system, but management and regu-

lation have an overarching influence on the distribution of the water and its quality. Nutrient en-

richment affects the rates of both aggregation and degradation processes, while fire can cause

soil degradation under certain conditions. The feedback loops are critical, as soil accretion or

loss alter future hydrologic conditions, and thereby future soil development.

The fresh water Everglades includes several soil-forming environments, differentiated primarily

on the basis of hydrologic regime and vegetation. As mentioned previously, in long hydroperiod

(8 to 12 months.yr-1) marshes, emergent graminoid or floating-leaved aquatic macrophytes are

the dominant primary producers, decomposition of roots and detrital materials is slow, and peats

are formed. In shorter hydroperiod (3 to 8 months.yr-1) prairies, macrophyte productivity is low-

er, and periphytic and benthic algal communities predominate (Davis et al. 2005; Ewe et al.

2006). High photosynthesis in the water column or in the benthos increases alkalinity, resulting

in the precipitation of calcium carbonate and contributing to the formation of mineral soils

(marls). Finally, the landscape also includes scattered forest fragments, or tree islands. Those

occupied by swamp forest trees are flooded for much of the year, and experience a soil develop-


87

ment process that in broad strokes resembles that of the marsh peatlands. Others occupy raised,

rarely flooded surfaces, and support “hardwood hammocks” with upland tree species. The soils

that form on them may be organic or mineral, depending on conditions.

Figure 2-4. Conceptual model of soil development in a managed wetland. Figure adapted from Ross et

al. (2006a).

Finally, it is important to note that soil development takes place within a continually changing

spatial context; that is, soil formation in one location is not independent of the landscape around

it. When the Everglades were first forming, the context was framed by the pitted bedrock sur-

face and the drainage network that connected the interior to the coastal estuaries. As soils

formed on this base, they smoothed out much of the fine-scale surface roughness, especially

where accretion was rapid. In other places, however, the spatial context was such that a pat-

Primary producers

Hydrology

Erosion

Degradation

Water quality

Nutrient enrichment

Water management

OM/calciteproduction OM

oxidation

Fire

Soil/Topography

Deposition

Aggradation

Regulatory agencies


88

terned landscape characterized by heterogeneity of an intermediate scale emerged. In this

“Ridge-and-Slough” landscape, the same processes illustrated in Figure 2-4 yielded a repeated

series of deeper and shallower organic soils, arranged in elongated landforms that parallel the

presumed, predominant long-term flow direction. Soil formation in the peatlands of the Ridge

and Slough are discussed further in the next section.

Peatlands

The hydrologic and nutrient linkages in Figure 2-4 are especially important in Everglades peat

soils, whose recent history has been a century-long story of subsidence in response to drainage.

Comparison of early land surveys to current elevations indicates that subsidence of 1-3 meters

has taken place in extensive areas south of Lake Okeechobee (Ingebritsen et al. 1999). Subsid-

ence in drained and tilled peat soils was first quantified by researchers at the USDA Belle Glade

station (Stephens and Johnson 1951). Soil loss in the EAA has slowed considerably from the

rapid rates characteristic of the first few decades of water management (Wright and Snyder

2009), but recent surveys indicate that significant soil loss has continued in some nearby portions

of the WCAs (Scheidt and Kalla 2007). As in other wetlands worldwide, subsidence of Ever-

glades peats increase with the depth of the water table. Data and modeling presented by Ste-

phens and Stewart (1977) indicated that peat subsidence increased as a linear function of depth

when the water table receded from 30 to 80 cm below the surface, with annual subsidence in-

creasing by about 7 mm per 10 cm increase in depth. The model did not predict changes in the

peat surface in flooded soils. Based on 137Cs activities in soil cores, Craft and Richardson (1993)

reported a mean accretion rate of 2.3 mm per year in un-enriched portions of WCA-2 and WCA-

3A. Fastest accretion was found in persistently flooded areas (lower WCA-3A, central WCA-

2A) and the slowest was found in over-drained areas (northern WCA-3A, WCA-2B). In the

same study, the most rapid peat accretion (~4 mm per year) occurred at a P-enriched site within

the cattail invasion front in WCA-2A. From the above, it appears that (1) decomposition ex-

ceeds production and the soil subside when water is well below the peat soil surface, while the

opposite is true when water rises toward and above the soil surface, and (2) phosphorus enrich-

ment may stimulate soil accretion, at least in the short run, perhaps by shifting the vegetation to-

ward the faster growing species.


89

Within the context of ridge and slough landscapes of the central and southern Everglades, the

influence of sedimentation and deposition (Figure 2-1) in peat soil dynamics has been an active

topic of recent research (Harvey et al. 2009; Larson and Harvey 2010; Leonard et al. 2006; Ba-

zante et al. 2006). While sediment redistribution may be an important process maintaining the

local balance between ridges and sloughs, there is no net change at the large scale, i.e., material

eroded from one location is likely to be deposited somewhere nearby. Furthermore, when the

landscape is in balance, accretion rates in sloughs should equal those in ridges (Cohen et al. un-

publ. data), though the respective rates may be arrived at by different combinations of produc-

tion, sedimentation/erosion, and decomposition.

The role of fire is an important element of any discussion of the dynamics of Everglades peat

soils (Figure 2-4). Newspaper accounts from Miami’s early years are filled with reports of un-

controlled fires in the wetland during the dry and early wet season (December to June), and there

is little doubt that drainage contributed greatly to the destructiveness of those fires. Craighead

(1971) reported major conflagrations in 1945, 1947, 1951-52, 1962, and 1965. Davis (1946) de-

scribed a site in which a 1945 fire burned enough peat to leave six inches of ash behind. Parker

(1974) described ash layers two inches thick, the remains of peat fires that burned to the water

table. These leading scientists contended that several south Florida soils, i.e., rocklands in the

East Everglades and sands in western Broward County, included extensive peat-coverage prior to

drainage. Even in the interior of the Everglades, fires burn across the surface during dry years,

but peat is consumed only when the water table has been artificially lowered by drainage, or dur-

ing very persistent or extreme climatic droughts. Unfortunately, no large-scale surveys of soil

loss have been initiated to date, leaving information on the interactions of hydrology, fire behav-

ior, and peat loss largely anecdotal at this point.

Marl Prairies

In Everglades marshes, marls are formed rather than peats when (1) vegetation production is low,

(2) production of calcareous periphyton is high, (3) a calcium carbonate source is abundantly

available, (4) the water column is well-lit, and (5) a significant dry-down occurs during most

years (Gleason et al. 1974; Browder et al. 1994). Marls do not form in P-enriched waters, be-


90

cause the blue-green algae that dominate calcareous periphyton mats are not tolerant of high P

(Gaiser et al. 2004). Once established and stabilized, they are not prone to entrainment by water,

as the silt particles that predominate are not easily dislodged. Depending on vegetation density,

fire frequency may be high in marl prairies, but their effects on soils may be minimal due to the

low soil organic matter content.

Marl soils are present on either side of Shark Slough in the southern Everglades, as well as in

fresh- and brackish-water prairies of the Southeast Saline Everglades, where they formed when

sea level was lower and freshwater runoff was higher. Today, those marshes are experiencing

mangrove encroachment, and formation of mangrove detritus based peat has ensued on top of the

marl base. The thickest south Florida marls are found in these prairies (Perrine series), with

depths sometimes approaching 1 m.

Relationships between marl accretion and hydrology, water quality, and other physical drivers

have not been quantified, though marl production is thought to be strongly associated with the

productivity of calcareous periphyton (Gleason 1972). Marl accretion is typically about 1 mm

per year (Meeder et al. 1996; Browder et al. 1994), which is only about one-half to one-third of

average rates for marsh peats. Nevertheless, sequestration of carbon in fresh water marls is sub-

stantial, due to their high bulk densities (Table 2-2). In fact, assuming the compiled values in

Table 2-2, annual estimates of carbon sequestered in peats (1290 kg.ha-1) and marls (1118 kg.ha-

1) in the Everglades are very similar.

Table 2-2. Physical data from several peat and marl soils.

Peat Marl

Mean annual accretion (mm) 2.31 1.02

Total carbon (%) 51.333 13.823

Dry bulk density (g/cm3) 0.113 0.813 1 Craft and Richardson 2008 (Cesium dating, WCA-2 and WCA-3) 2 Meeder et al. 1996 (Cs and Pb dating, C-111 basin, Perrine Series) 3Ross et al. unpublished data (southern Shark Slough)


91

High Tree Islands

Hardwood hammocks in the southern Everglades grow on two distinct soil types (Ross et al. un-

publ. manuscript). The first type is common in the seasonally flooded marl prairie landscape, and

consists of shallow, organic, relatively low-P soils formed directly on limestone outcrops. In

contrast, hammocks on islands embedded in long hydroperiod marsh have deeper, alkaline, min-

eral soils with extremely high P concentrations. The first group resembles the organic soils

found in Florida Keys hammocks, where soil development consists primarily of a reprocessing of

dead roots and remains of aboveground plant tissue. Maintenance of these soils requires sustain-

ing high aboveground production, because decomposition rates in such well-aerated settings are

rapid. Development of the second type of tree island soil, which is restricted to islands envel-

oped throughout the year by flooded or saturated marsh peats, is not yet completely understood.

Several explanations for their high mineral content have been proffered. One is that the mineral

component is the residuum of rapid organic matter decomposition and weathering of the bed-

rock. A second is that the source is bone material transported to the islands by aboriginal occu-

pants or visitors, and which suffuse many of these profiles. A third hypothesis is that the miner-

als are fixed through subsurface precipitation of calcium and other cations, drawn to the islands

in the transpiration stream of trees. Some support for this mechanism is provided by Wetzel et

al. (2011), who reported high Na and Cl ionic concentrations in groundwater beneath an elevated

tree island in WCA-3A. Each of the above mechanisms probably contributes to soil develop-

ment in high islands in interior peatlands.

Compared to marsh peats, the well-drained soils of these upland inclusions would not be espe-

cially sensitive to enhanced decomposition rates associated with reduced water tables (Figure 2-

4). However, the organic soils in the marl prairie tree islands are very vulnerable to fire, whose

effects can be exacerbated by reductions in seasonal water levels. In contrast, the mineral soils

in the high slough islands probably maintain more stable moisture conditions due to supplemen-

tation from adjacent marsh waters, and would not be consumed by the rare fire that reaches them.

Prolonged high water could have indirect effects on soils in either setting, via its influence on the

forest canopy. Soil dynamics in both types are apparently dependent on a continuous tree layer

which contributes organic materials, draws water and nutrients into the islands through the tran-


92

spiration stream, attracts animals, intercepts aerosols, and controls the belowground microcli-

mate. Flooding has eliminated forest canopies in some parts of the Everglades (Hofmockel et al.

2008), and changes in soils are now under investigation (Ewe et al. 2010).

Human impacts to soils _____________________________________________________________________________

Nutrient Loading Impacts

Beginning with the first attempts to drain the northern Everglades to bring the rich organic soils

south of Lake Okeechobee into agricultural production, humans have altered the unique envi-

ronmental conditions that supported the ecosystem and the accretion of large expanses of histo-

sols (organic soils) found there (Davis 1994). With the advent of drainage within the EAA circa

1900-1910, subsidence of organic soils began. Beginning in 1913, extensive soil subsidence

within the EAA was investigated and documented (Davis 1943; Jones 1948; Stephens and John-

son, 1951; Gleason and Stone 1994; Snyder and Davidson 1994; Snyder 2005). By draining the

waterlogged soils, a century of soil oxidation was catalyzed that has resulted in not only the great

loss of soil (up to 3m in some areas) in the upper Everglades basin (Everglades Agriculture Area

- EAA), but has also contributed, by way of organic matter mineralization, to elevated nutrient

runoff to the remaining Everglades downstream (WCAs, ENP). Further, liberal use of agro-

chemicals and fertilizers has contributed to increased presence of xenobiotics and nutrients in

waters flowing into the Everglades today (SFWMD 2007), requiring the use of extensive storm-

water treatment areas or STAs—treatment marshes designed to remove and store nutrients from

agricultural waters prior to release into the environment (Pietro et al. 2007). About 50% or more

of the nutrient laden water from the EAA is diverted to the St. Lucie and Caloosahatchee Estuar-

ies via canals (Redfield and Efron 2007; Richardson and Huvane 2008). The remaining portion

enters the Everglades through a series of drainage canals and water control structures. These in-

puts of agricultural runoff carry an assortment of metals, organic chemicals, and nutrients that

are detrimental to the system. The most ecologically notable is phosphorus (Davis 1991; Craft

and Richardson 1993; Reddy et al. 1993).


93

Another factor that has significantly impacted the Everglades is the rapid increase in human pop-

ulation and the concomitant urban sprawl. People have lived in the Everglades region for more

than a thousand years, however within the last century, their activities profoundly impacted this

ecosystem. The Everglades have been reduced to roughly half the original spatial extent by ag-

gressive encroachments from both residential and agricultural land uses (Walker 2001). The

ecosystem has been under continuing threat from increasing population density (Walker et al.

1997). Currently, the threat of unchecked urban sprawl has been curtailed due to the federally

mandated Urban Growth Boundary (UGB), a line establishing the urban development limits to

prevent any more encroachment on the Everglades.

With the population and metropolitan growth came urban problems such as overloaded sewage

treatment plants and water shortages. Wilcox et al. (2004) demonstrated over 60% of the water

being removed by municipal pumping in Miami-Dade county originated in the Everglades.

Meanwhile, over the past three decades, the drainage canals conveying urban and agricultural

runoff to the Everglades contained water with high total phosphorus concentration ranging from

100-1000 µg L-1 (Sklar et al. 2005), orders of magnitude in excess of CERP goals.

The 1970’s through 1990’s witnessed significant ecosystem decline seemingly linked to phos-

phorus enrichment. This resulted in much attention and research concerning P loading to the

WCAs and subsequent ecosystem changes such as vegetative community shifts and habitat deg-

radation that were attributed in the most part to excess soil phosphorus (SFWMD 1992; Davis

1994; Noe et al. 2001; McCormick et al. 2002). Due to overwhelming evidence of P enrichment

in the northern Everglades, several studies have been conducted to investigate P enrichment in

soils (Koch and Reddy 1992; Amador and Jones 1993; Craft and Richardson 1993; Reddy et al.

1993; DeBusk et al. 1994; Qualls and Richardson 1995; Amador and Jones 1995; Newman et al.

1996, 1997, and 1998; Miao and Sklar 1998; Noe et al. 2002; Noe et al. 2003; Daoust and Chil-

ders 2004; Chambers and Pederson 2006) and in some cases, changes to soil condition over time

(Childers et al. 2003; Reddy et al. 2005; Bruland et al. 2006; Rivero et al. 2009; Newman et al. in

press). These studies overwhelmingly contend that inflow waters from agriculture operations

upstream have caused P enrichment of soils in many areas across the Everglades landscape

(Scheidt and Kalla 2007; Hagerthey et al. 2008; Osborne et al. 2011a).


94

As mentioned previously, the Everglades was historically an oligotrophic system, and as such,

the availability of essential nutrients such as P was very limited. Other than internal recycling of

P from soils to plants, sources of P inputs were few, with atmospheric deposition being the most

significant source (Scheidt et al. 2000; Noe et al. 2001). The limited availability of P translates

to very tight biogeochemical cycling of P forms in the ecosystem, leaving bioavailable forms in

very low concentrations and a majority of the P sequestered in plant or microbial tissues and par-

ticulate detrital material (Figure 2-5). This nutrient limitation brought about ecosystem domi-

nance of plant communities adapted to growing in low P conditions (Davis 1991; Miao and De-

Busk 1999; Miao and Sklar 1998).

A majority of studies and the scientific community working in the Everglades concur that even

small additions of P can have a dramatic effect on ecosystem productivity and functioning

(Gaiser et al. 2005; Childers et al. 2001; Chiang et al. 2000). Nowhere are these effects more

readily observed than in the changes to vegetation. Autochthonous nutrient inputs have resulted

in significant alterations to the indigenous system with large expansions of cattail (Typha

domingensis; Davis 1994; Newman et al. 1998; Richardson et al. 2008). Cattails are adapted to

grow rapidly in the presence of available P, and in doing so, out-compete native vegetation such

as sawgrass (C. jamaicense; Davis 1991; Davis 1994; Miao and Sklar 1998; Miao and DeBusk

1999).

These changes to vegetation have far reaching consequences to the ecosystem. Beyond the visu-

al changes, changes in vegetation result in changes to the heterotrophic food web by way of al-

tered organic matter quality. This also affects soil accretion processes mentioned previously.

Accelerated decomposition of cattail detritus over sawgrass has been observed (Craft and Rich-

ardson 1997; DeBusk and Reddy 1998) suggesting rapid turnover of soil building materials.

Phosphorus re-absorption during senescence has also been found to be less efficient in cattails

vs. sawgrass, resulting in greater amounts of P being deposited to the detrital pool (Miao and

Sklar 1998; Osborne et al. 2007). Further, rapid organic matter decomposition due to lower lig-

nin content of cattails also increases the turnover rate of P, resulting in a positive feedback

mechanism to encourage Typha expansion. Other ecological implications of cattail expansion

include changes to water quality via dissolved oxygen content, fish and wading bird habitat deg-


95

radation, accelerated biogeochemical cycling of nutrients and metals of concern, such as mercury

(Osborne et al. 2011a), and dramatic changes to the calcareous periphyton communities (Gaiser

et al. 2005; Gaiser et al. 2006).

As with sawgrass and other low P adapted plants of the Everglades, the unique calcareous pe-

riphyton found almost ubiquitously throughout the system is adapted to low P availability

(McCormick and Stevenson 1998). Periphyton is composed of filamentous cyanobacteria and

diatoms assembled together in a laminar sheet that can be attached to the soil surface, plant sur-

faces below the water column, or floating on the water surface (Figure 2-6). During the process

of marl formation (calcium carbonate precipitation), P is often co-precipitated, becoming part of

the soil (McCormick et al. 2001; Gleason et al. 1974). Periphyton is indicative of pristine condi-

tions and is an integral part of ecosystem functioning in that it is responsible for significant pri-

mary production, P sequestration, carbon sequestration, and actively regulates other biogeochem-

ical processes in the water column (McCormick and O’Dell 1996; Noe et al. 2001; Gaiser et al.

2006). When water column P concentrations exceed approximately 20 ug l-1, cyanobacteria are

replaced by filamentous green algae (McCormick et al. 2001; Gaiser et al. 2005 and 2006). The

effect on marl formation and accretion by way of the loss of calcium carbonate precipitating cy-

anobacterial component is detrimental.

In the Everglades surface water quality and vegetative communities are intricately inter-related

with soil characteristics (Daoust and Childers 2004; Davis et al. 2005; Ogden 2005; Hagerthey et

al. 2008). Diminished surface water quality has resulted in altered vegetation and periphyton

communities, which have in turn altered litter quality, microbial activity (Wright and Reddy

2001), mineral precipitation, and ultimately soil accretion. These altered plant communities may

cause further changes in soil type and thickness as these different plant communities eventually

decompose and form altered soil (Scheidt and Kalla 2007). Of considerable concern is the re-

duction of C storage and soil accretion, reduction of P storage, and resulting increase of available

P. The later serves as a significant positive feedback for further vegetation and periphyton

changes and increased microbial decomposition of litter and soils.


96

Figure 2-5. Phosphorus cycle in Everglades marsh. Abbreviations are as follows: dissolved organic phos

phorus (DOP), dissolved inorganic phosphorus (DIP), particulate organic phosphorus (POP), particulate inorganic phosphorus (PIP) and inorganic phosphorus (IP). Artwork by P. Inglett from Reddy and DeLaune (2008).

Figure 2-6. Schematic depicting three levels of periphyton colonization and phosphorus cycling by

periphyton including co-precipitation of calcium bound phosphorus. Artwork by P. Inglett from Reddy and DeLaune (2008).


97

Figure 2-7 represents the total phosphorus content of the soil in the Everglades calculated on

mass (mg P kg-1) and volume basis (µg P cm-3) (Scheidt and Kalla 2007). This figure indicates

that in 2005, 24.5 ± 6.4% and 49.3 ± 7.1% of the Everglades area soil phosphorus content ex-

ceeded 500 mg kg-1 and 400 mg kg-1, respectively (Figure 2-7). These numbers are to be com-

pared with those obtained in 1995-1996 (16.3 ± 4.1% exceeding 500 mg kg-1 and 33.7 ± 5.4 %

exceeding 400 mg kg-1). The Everglades areas with the highest soil TP concentration are gener-

ally the peat soils located in WCA-3A north of Alligator Alley, northern WCA-2A, and the edg-

es of LNWR close to the rim canal (Figure 2-7). When expressed on a volume basis (µg cm-3) to

differentiate among the soil types, the peat soils with higher TP content are located in WCA-2A

and at the edges of LNWR. The areas in the ENP with higher bulk density have a higher P con-

tent. The locations in southern WCA-3A that contained above 500 mg kg-1 (mass basis) have a

low TP content on a volume basis.

Figure 2-7. Total phosphorus content in soil expressed on a mass basis (mg kg-1) and on a volume basis

(µg/ cubic centimeter or µg cm-3) (Scheidt and Kalla 2007).


98

Impacts Associated with Addition of Terminal Electron Acceptors (TEAs)

In addition to nutrient loading, human impacts to Everglades soils can also occur through the in-

creased availability of alternate terminal electron acceptors (TEAs), which replace oxygen for

respiring microorganisms in the soil under anaerobic conditions. The addition of new or alter-

nate TEAs can be derived from direct anthropogenic sources (e.g., nitrate inputs from agricultur-

al sources) as well as through saltwater intrusion near oligohaline coastal zones (e.g., sulfate as-

sociated with seawater inundation) and can result in increased soil respiration rates thereby af-

fecting the stability of soil carbon reserves and, if impacts are sustained over long periods of

time, soil subsidence can occur.

Typical of wetland soils, oxygen is rapidly depleted from wetted Everglades soil, as the rate of

oxygen consumption is much greater than the rate of supply under saturated or inundated condi-

tions (Mitsch and Gosselink 2000). Oxygen is preferentially consumed because oxic respiration

yields the greatest amount of energy to sustain life. Following oxygen depletion, there is a rela-

tively predictable sequence of respiratory pathways leading to TEA depletion, which is related to

redox state and ultimately tied back to energy yield (Reddy et al. 1986; Reddy and D’Angelo

1996). However, TEA availability alone does not necessarily dictate soil respiratory pathways in

peat wetlands (Dettling et al. 2006).

Nitrate (NO3-) and sulfate (SO4

-) are the essential TEAs in fueling denitrification and sulfate re-

duction, respectively. Denitrification is a respiratory pathway that is energetically favored as

oxygen becomes depleted (around 250 mV soil redox) and there is a sufficient supply of organic

substrate and nitrate (NO3-)—an oxidized inorganic form of nitrogen. This pathway results in

the reduction of nitrate ultimately to di-nitrogen gas (N2), as organic matter is oxidized (Reddy

and DeLaune 2008). Sulfate, an oxidized inorganic form of sulfur, is used as a TEA for sulfate

reduction at redox levels around -100 to -200 mV, resulting in reduction of sulfate to sulfide (S2-)

(Reddy and DeLaune 2008). Other wetland studies (e.g., Neubauer et al. 2008) have shown Fe

III to be important in anaerobic soil respiration. Little is known about the significance of iron

reduction in Everglades soils, although Qualls et al. (2001) indicate reduction of iron when Fe III

is more available. This could be a significant respiratory process to consider in impacted or nat-


99

urally Fe-rich areas of the Everglades, particularly as it is associated with the dissolution and

mobility of soil phosphorus (e.g., Aldous et al. 2005).

Although these pathways commonly occur in Everglades soils, the enrichment of TEAs such as

nitrate and sulfate in the Everglades soil environment can result in accelerated C mineralization

and concomitant nutrient or contaminant remobilization. The former is especially important giv-

en the amount of fixed carbon stored in Everglades peat soils and the importance of maintaining

a positive balance between peat soil production and decomposition. The latter is critical to con-

trolling water quality impacts, as Everglades soils represent the largest reservoir of nutrients

(particularly P) within the ecosystem. In addition to being a source of nutrients, enhanced peat

decomposition has the potential for a positive feedback effect on the availability of TEAs. As

peat is oxidized, soil-bound sulfur and nitrogen can be mineralized and either retained by the soil

or transmitted downstream via surface or groundwater.

Nitrate in wetlands is largely derived from precipitation or the chemoautotrophic oxidation of

ammonium to nitrite and ultimately to nitrate that occurs at the interface between reduced wet-

land soils and the more oxidized water column or at rhizospheric interface. In a natural state, the

source of ammonium is from the breakdown of organic matter that leads to the mineralization of

nitrogen (ammonification). In impacted areas of the Everglades, inorganic nitrogen is derived

from canals or through enhanced soil N mineralization in areas of shortened hydroperiod.

Compared to freshwater ecosystems, sulfate is often much higher in brackish or saline waters.

Therefore, sulfate reduction is often a dominant biogeochemical pathway in coastal wetlands

worldwide, including the mangrove swamps of the coastal Everglades. The major source of sul-

fate contamination in the Everglades appears to be discharges from canals originating from EAA

soil oxidation and agricultural sulfate applications as a fertilizer. In fact, Bates et al. (2002)

showed that the 34S content of water coming from the EAA reflects that of agricultural fertilizer,

and that ground water under the EAA may also be a significant contributing source. Other man-

made water bodies that pierce deeper, more saline parts of the aquifer may also be a contributing

source (Bates et al. 2002). When other, more energetically favorable TEAs have been depleted,

a model by Lovely and Klug (1986) suggests that a sulfate reduction zone should develop at


100

SO42- levels greater than 30 µm due to their ability to keep acetate levels low enough to limit

methanogenic activity. At levels below this, a methanogenesis zone should develop (Lovely and

Klug 1986).

Impacts from Modified Hydrology

The change in Everglades hydrology over the last century has been pondered and documented by

many (Light and Dineen 1994; Steinman et al. 2002), and the biological changes that ensued are

being unraveled through paleoecological studies and recent time-series analyses of vegetation

and faunal changes (McIvor et al. 1994; Sklar et al. 2002a; Willard and Cronin 2007; Bernhardt

and Willard 2009), some of which are linked to phosphorus loading (McCormick et al. 2002;

Craft and Richardson 2008). From these studies, we know that impacts to soils have been signif-

icant and are related primarily to changes in soil redox (linked to altered local hydroperiod), the

loss of soil organic matter in areas that have received less water, and an increase in soil organic

matter (i.e., peat accumulation) for areas that have been exposed to increased hydroperiod—

especially in combination with increased P loading (Craft and Richardson 2008). DeBusk and

Reddy (2003) showed very clearly the relationship between declining water levels in the soil and

C flux—a relationship that was enhanced with P enrichment. And an extreme example of the

impacts of modified hydroperiod on soil organic matter has been documented in the drained soils

of the EAA, where soil subsidence rates have been measured on the order of several meters over

the past century (Lodge 2010). In relatively un-impacted tree islands, Troxler-Gann and Chil-

ders (2006) showed that much of the variability in ecosystem structure of tree islands was ex-

plained by soil oxidation state and hydroperiod.

In terms of the effects of modifying hydrology on nutrient dynamics, Corstanje and Reddy

(2004) illustrated that re-flooding disturbed soils in central Florida can release soil P and N (as

phosphate and ammonium) to floodwater. A similar study by Aldous et al. (2005) indicated that

the effect of increased hydroperiod in restored wetland soils may exhibit an increased mobility of

P in reduced, iron-rich soils. This is unlikely to be a problem in highly organic peat soils or car-

bonate-rich marl soils of the Everglades. However, studies have shown that soils that had been

dried also exhibited significant P efflux upon re-wetting due to the oxidation of organic soils


101

(Pant and Reddy 2003; Aldous et al. 2005). Lastly, White and Reddy (2003) concluded that wa-

ter management was critical in maintaining the coupling of nitrification and denitrification,

which to a great extent regulates the soil N balance.

In addition to peat desiccation and soil subsidence, a more rapid and serious threat to peat soil

stability in areas that have experienced reduced hydroperiod is fire. The Everglades fire regime

has changed with drainage and subsequent water management (Lodge 2010). During very dry

years or following consecutive years of low rainfall, severe fires sometimes result in the combus-

tion of organic rich soils that have developed over hundreds of years—leading to changes in lo-

cal hydroperiod and plant community (Lodge 2010). Muck fires also shift the balance of soil P

pools from less mobile, more organic forms of P to more mobile, less organic P, which can have

direct enrichment effects to the water column upon re-flooding (Leeds et al. 2009).

In areas dominated by marl soils (e.g., western and eastern ENP), modified hydrology can influ-

ence periphyton composition and production, which can both affect the edibility and accretion of

periphyton mat material into the soil (see review by Davis et al. 2005). Also, see Gottlieb et al.

(2005) for experimental result of hydroperiod manipulation on periphyton species and chemical

composition, location/position, and biomass.

Geographic Areas of Concern

Several issues pertaining to soils at a landscape scale identified in the literature reviewed here

give reason for concern. Extensive areas in LNWR continue to be impacted by P and S enrich-

ment even in light of recent improvements to water quality from the EAA. Similarly, WCA-2A

appears to have some stabilization in soil P enrichment along the eutrophic gradient in the north-

ern portion of the unit; however, recent spatial and temporal comparisons indicate translocation

of TP and suggest possible internal cycling as mechanisms for increased spatial extent of P im-

pact to soils in light of overall reduced soil P concentrations (Marchant et al. 2009; Reddy et al.

2011), as well as, continued water quality inputs in excess of CERP goals. Sulfur enrichment in

WCA-2A, combined with current sulfide distribution in surface soils suggests that extensive sul-


102

fate reduction is occurring in the eutrophic areas, further complicating P mobilization problems

(Osborne et al. 2011a).

Of great concern is the extensive soil subsidence in WCA-3A, as described by Scheidt and Kalla

(2007). Landscape trends suggest that if this subsidence is not mitigated hydrologically, soil

losses will continue. Soil oxidation in the Everglades has great impact on the surrounding land-

scape as nutrients and contaminants stored in these peat deposits will be mobilized causing fur-

ther exacerbation of eutrophication and contamination downstream. As P enrichment is signifi-

cant in this area, subsidence here has great potential to increase local enrichment, along with the

resultant expansion of cattail, and increase P export / translocation as well. Spatial patterns of S

and mercury (Hg) suggest extensive accumulation of both of these potentially detrimental ele-

ments in soils south of the most severe soil subsidence (Scheidt and Kalla 2007; Reddy et al.

2005). This could be an effect of previous subsidence and mobilization; however, the most rele-

vant threat is remobilization via soil oxidation in these areas of concentration.

Positive findings of minimal P enrichment in southern WCA-3 and ENP give credence to the

prioritizing of these areas to protect from future encroachment by excessive nutrients and con-

taminants. However, while soil concentrations of P are low in the ENP, recent evidence suggests

accelerated P enrichment in Taylor Slough, a key habitat area of ENP (Reddy et al. 2008). This

finding suggests that vigilance in soil resource protection is required to maintain the relatively

unimpacted areas in their present condition while concomitantly working to restore the enriched

areas.

Lastly, the hydrologic impacts on tree island soils, particularly in WCA-3A and 3B, and the po-

tential impacts of sea-level rise/saltwater intrusion on soils at the oligohaline interface cannot be

overlooked. However, these issues associated with soil reserves have not been sufficiently doc-

umented or vary over relatively small spatial scales.


103

Management implications of Everglades restoration on soils ______________________________________________________________________________

Restoration Mandates for Everglades Soils – CERP and MAP

Restoration Coordination and Verification (RECOVER) has developed performance measures

for evaluating modeled system-wide performance and assessing actual system wide performance

of CERP in meeting its goals and objectives. These performance measures are tools based on a

set of indicators developed through conceptual ecological models (CEM) that identified key

stressors and attributes of the natural system. These CEM can be found in the Monitoring and

Assessment Plan (MAP), the primary tool by which the RECOVER program assesses the per-

formance of the CERP (or Plan). The 2009 RECOVER MAP (RECOVER 2009) is an updated

version of MAP 2004 describing 1) the monitoring components and supporting research of the

MAP 2) summarizing the assessment process and 3) developing the conceptual ecological mod-

els necessary to establish the system wide performance measures. The scientific and technical

information in the MAP allow RECOVER to assess CERP status and performance.

Soil is recognized as an important component in the functioning of the Everglades ecosystem in

the CERP system-wide MAP. This program includes performance measures that address soil

condition directly (phosphorus in soil, soil loss) or indirectly (water inundation pattern, drought)

as soil is impacted by other stressors, however, MAP does not presently provide for soil

condition monitoring.

Phosphorus

Eutrophication of extensive areas of the northern Everglades via P-laden runoff from the EAA

and the subsequent visual changes to the landscape was one of the significant catalysts in the res-

toration movement (Davis and Ogden 1994). The impact of elevated P to the system is consid-

ered so detrimental it is included in several conceptual ecological models and performance

measures used in framing restoration planning (Ogden 2005; RECOVER 2006). The degrada-

tion of the northern Everglades marshes due to P enrichment is extremely well documented

(Scheidt et al. 2000; Noe et al. 2001; McCormick et al. 2002; Childers et al. 2003; Hagerthey et


104

al. 2008), and as such, significant scientific evidence exists to aid restoration efforts with respect

to P.

Elevated soil nutrient concentrations, attributed to anthropogenic activities, have resulted in sig-

nificant shifts in the nutrient sensitive biological communities in the oligotrophic Everglades.

Soil TP concentrations of between 500 and 600 mg kg-1 have frequently been used by research-

ers to indicate areas of enrichment in the Everglades (Craft and Richardson 1993; Reddy et al.

1991). Other studies have indicated that soil P concentrations of 650 mg kg-1 facilitate the ex-

pansion of cattails (Typha spp.) resulting in the loss of native marsh communities dominated by

sawgrass (Cladium jamaicense) (Wu et al. 1997). Florida’s Everglades total phosphorus criteri-

on rule specifies a definition of impacted as being where soil TP exceeds 500 mg kg-1 of soil and

CERP maintains a restoration goal of 400 mg kg-1. Indeed, based on extensive analyses con-

ducted by FDEP, sediment P concentrations in the 500 to 600 mg kg-1 range appear to corre-

spond to areas exhibiting long-term water column TP concentrations above 10 ppb (Payne et al.

2003). In other words, the application of a 10 ppb TP criterion for the water column, which is

above background concentrations of 5-8 ppb, will maintain a long-term geometric mean TP con-

centration in soil at slightly above background levels of 200 to 400 mg kg-1.

Total phosphorus concentration in soil is an effective means to evaluate long-term ecosystem

impacts and is crucial to assessment of CERP activities. Among the MAP working hypotheses

used to establish and assess the soil nutrient performance measure were:

1) Increased phosphorus concentrations and loads in agricultural runoff water, and replacement

of sheet flow with canal flows and point-source discharges, have produced phosphorus concen-

tration gradients in surface water, soil and periphyton downstream of canal discharge structures

and shifted wetlands from oligotrophic to eutrophic states (McCormick et al. 1996);

2) Research has shown that assessment of water column nutrient concentrations alone does not

generally capture the level of ecosystem impact (restorative or deconstructive) that results from

changes in nutrient loads over a period of time (Gaiser et al. 2004; Reddy et al. 2008);


105

3) Soil nutrient concentration is a long-term indicator of nutrient loading that in the Everglades is

often associated with ecosystem change (Scheidt and Kalla 2007).

The performance target set by CERP is to decrease the areal extent of TP concentrations

exceeding 500 mg kg-1 and maintain and reduce long-term average concentrations to 400 mg kg-1

or less in the upper 10 cm of soil (RECOVER 2007). Based on analysis of available data and

discussions among SFWMD, the United States Environmental Protection Agency (USEPA), and

the United States Fish and Wildlife Service (USFWS) scientists, a numeric sediment TP

concentration target in the range of 200 to 400 mg kg-1 in the top 10 cm of soil column, including

the overlying layer of flocculent or unconsolidated sediment, will be used to delineate areas

minimally impacted (biologically) from those in which the structure and function of the native

biological communities have been significantly altered by P enrichment. This target is variable

in space and only applicable to peat soils (RECOVER 2007).

Scheidt and Kalla (2007) reported that during 2005 soil phosphorus exceeded 500 mg kg-1, Flor-

ida’s definition of “impacted”, in 24 % of the Everglades, and it exceeded 400 mg kg-1, CERP’s

restoration goal, in 49 % of the Everglades. These proportions are higher than the 16 % and 34

%, respectively, observed in 1995-1996.

Carbon

Drainage of the ecosystem, compartmentalization and reduction of the total water quantity stored

in the ecosystem has exaggerated the dry seasons and dry years that can follow. It has been ob-

served throughout the northern Everglades that peat loss is associated with changes in water de-

liveries that reduce water depth and hydroperiod duration. The result is increased rates of organ-

ic soil loss (Sklar et al. 2000) that can occur as either a multi-decadal impact associated with

gradual shifts in the relationship between organic matter accumulation and soil respiration

caused by changes in hydropattern, or as an immediate effect of rare and intense peat soil con-

suming fires. The CERP establishes a performance measure that addresses extreme low and high

water levels and ultimately benefits Everglades soils. The performance measure examines the

frequency, duration, and percent period of record of extreme events and the peat exposure due to


106

droughts (RECOVER 2007). The intent of the drought intensity component of the performance

measure is to use a quantitative graphical display of cumulative desiccation intensity (magnitude

and duration of drying event) to determine whether alternative project designs are likely to in-

crease or decrease the potential for further unnatural loss of organic soils. The goal is to reduce

the risk of further loss of soil elevations due to excessive drying.

A second performance measure related to soil preservation is the restoration of the sheet flow in

the Everglades ridge and slough landscape to re-establish the natural patterns of distribution, tim-

ing, continuity and volume of sheet flow (RECOVER 2007). This will significantly help to sus-

tain the microtopography in relation to organic soil accretion and loss (sheet flow interacts with

hydroperiod, water depth, fire, and nutrient dynamics to maintain organic soil accretion and loss

in a state of dynamic equilibrium). This performance measure will be applied to a set of tran-

sects near Tamiami Trail and within the ENP including central Shark Slough and Taylor Slough.

Timing and distribution index scores are introduced and provide information about how the tim-

ing of discharges across transects and flow distribution across individual transects are altered by

alternative project configurations (RECOVER 2007).

Other performance measures like dry events in Shark River Slough or the inundation patterns in

the Everglades wetlands will also help restore the soil (formation and maintenance) in the Ever-

glades since the soil peat accretion typical of the ridge and slough landscape requires prolonged

flooding, characterized by 10 to 12 month annual hydroperiods, and groundwater that rarely

drops more than one foot below ground surface.

Approaches to Assessment of Everglades Soils

To assess the effectiveness of restoration projects, extensive monitoring of soil along with other

parameters is required. Regular sampling to determine the spatial distribution of soil physico-

chemical characteristics is important to determine the long-term impacts and changes. Several

monitoring design options are currently available. These options should be carefully assessed as

to their advantages and disadvantages prior to selecting a design.


107

Spatial

The Everglades Protection Area (EPA) is a spatially large and diverse ecosystem. It is important

to monitor the system as a whole versus select areas to determine and follow the changes in soil

components (Scheidt and Kalla 2007). A randomized design, similar to the one developed in the

R-EMAP study is capable of spatially integrating broad-scale changes for the entire landscape

area. The R-EMAP design is utilizing a probability based sampling design (reviewed and rec-

ommended by NRC) with a total of 1145 sites. Each iteration of system wide sampling included

hundreds of sample locations across the EPA, enabling temporal as well as, spatial interpretation

of patterns and trends in soil attributes. The preeminent strength of the probability-based design

used by R-EMAP is the ability to make quantitative statements across space about the status of

indicators of ecological health with known confidence limits across space. This design is desira-

ble if a broad spatial integration is the highest priority (Scheidt and Kalla 2007). Another strati-

fied-random sampling method was selected by Reddy et al. (2005) in conjunction with the

SFWMD and the RECOVER Assessment Team to conduct a system wide soil sampling effort,

the Everglades Soil Mapping project (ESM) at 1358 sites in a single phase. This sampling de-

sign was based on the one implemented in LNWR in the 1990s (Newman et al. 1997) to ensure

that large zones with low variability in physico-chemical soil properties were not over-sampled

and small zones with high variability were not under-sampled (DeBusk et al. 2001; Grunwald et

al. 2008). This will guarantee that higher sampling density for areas previously characterized by

high spatial variability of soil P concentration (Richardson et al. 1990).

Transects/Sentinels/Primary Sampling Units

Transects are an excellent approach toward assessing changes along gradients. These gradients

are the most likely locations where ecological change will occur; transects maximize the ability

to detect this change. If a transect design is chosen, sampling along gradients can either be com-

pletely randomized or stratified random sampling (depending on steepness of gradient or on hab-

itat type); sample sites can be fixed for some parameters (with randomized initial selection) such

as for ground water wells or individual trees (Scheidt and Kalla 2007; Reddy et al. 1998).

Another stratified random sampling design is based on primary sampling units (PSU). In this

design, the greater Everglades is divided into a series of landscape subunits (LSU). Within each


108

subunit, the design recommends selection of PSUs that be selected based upon availability and

location of sparse emergent freshwater marsh within each landscape subunit. After the PSUs are

selected, sampling locations will be randomly chosen within the appropriate habitat of each PSU.

The PSU design is currently used by several researchers (UF and FIU) in the MAP and RE-

COVER monitoring of tree islands and ridge slough status. A recent study of short range spatial

variability with respect to soil P also utilized this study design (Cohen et al. 2009).

Another design was also suggested for sentinel site sampling. Sentinel sites will be located to be

representative of regionally significant ecotypes and intended to provide information for regions

of interest either in impact areas or in reference, un-impacted regions. These sentinel sites could

be existing fixed station monitoring sites with historical data.

While there are several alternative spatial sampling design options, the utilization of any one in

particular is objective driven as there are substantial arguments for the use of each one. The most

pressing management question with respect to soil monitoring currently, is how often this type of

sampling should occur. R-EMAP was projected to occur in phases that spanned 3-5 years. The

last phase was completed in 2005 and at this time, there are no plans to continue this effort. The

intent of ESM, from the standpoint of RECOVER, was to complete this system wide sampling

on a 7-10 year cycle. Smaller scale sampling efforts are definitely more financially feasible while

landscape scale monitoring is very resource demanding. However, several studies contend that

this level of sampling, at some set temporal interval, is critical for monitoring ecosystem health

and assessing restoration success (Scheidt and Kalla 2007; Osborne et al. 2011a).

In-situ experiments

Field or in-situ experimental design is another option to be selected when quantification of eco-

logical effects of chemical inputs is needed. Field studies which enclose a portion of the natural

system of interest for experimentation offer a promising compromise between small scale labora-

tory and large scale, uncontrolled natural experiments (Richardson et al. 1992). These systems

allow for ecological responses to specific perturbations to be quantified under ambient exposure

conditions (water flow, temperature fluctuations) while providing a degree of experimental con-

trol and replication rarely obtainable in the unconfined ecosystem (McCormick et al. 1994).


109

These characteristics reduce problems associated with extrapolating experimental findings to the

natural system and allow for ecological responses to phosphorus to be isolated from other envi-

ronmental influences. While variability among replicate mesocosms is generally higher than for

laboratory test systems, such problems can be minimized by replicating treatments sufficiently to

allow for an acceptable level of statistical power. As an example, Sklar et al. (1993) implement-

ed a field monitoring study to document the nature and extent of ecological impacts associated

with gradient of phosphorus enrichment in the Everglades (Hagerthey et al. 2008).

Measurements of Sedimentation and Erosion

Changes in soil elevation summarize the balance between elevation gains due to accretion or

deposition of material and the losses due to shallow subsidence, compaction, and/or erosion

(Whelan et al. 2009). The combined use of soil elevation tables (SETs) and feldspar marker

horizons provide an integrated measure of elevation (i.e., deposition minus subsidence). The

SET can be used to determine both the influence of a single meteorological event on sediment

surface elevation and a long-term trend (i.e., decades) in elevation change (Cahoon et al. 2002a).

Figure 2-8 presents the original SET device (RECOVER 2006). The SET consists of a

mechanical arm that is attached to a benchmark and leveled, establishing a fixed measuring

point. Typically each SET has four fixed measurement locations (directions), where nine

measuring pins are lowered to the soil surface to obtain a relative soil elevation. The elevation is

the mean of 36 measuring pin readings per benchmark.

A new portable mechanical leveling device (Rod Surface Elevation Tables-RSET) was devel-

oped for high-precision measurements of sediment elevation in emergent and shallow-water wet-

land systems. It works on the same principle as the SET. However, the new device is an im-

provement in the determination of elevation change occurring over different parts of the sedi-

ment profile because it can be attached to benchmarks that are driven to both deeper and shal-

lower depths than the SET. Cahoon et al. (2002b) provides descriptions and several detailed di-

agrams of the RSET and the deep (driven to refusal) and shallow (< 1 m depth) stable bench-

marks to which it can be attached.


110

Figure 2-8: Conceptual illustration showing the use of SETs and feldspar markers to measure soil

elevation through time.

Available Data

Early soil surveys in the Everglades (Davis 1946, John 1948) provided the baseline to understand

physical properties of dominant soils (Histosols) and to document changes from subsidence of

organic soils caused by extensive drainage occurring in the EAA between 1910 and 1920 (Ste-

phens and Johnson, 1951; Gleason and Stone 1994; Snyder and Davidson 1994; Snyder 2005).

During the last 15 years and as a result of agriculture activities occurring in the EAA and in Lake

Okeechobee watershed, soil studies in the Everglades shifted their focus to examine soil biogeo-

chemistry and P enrichment. During the 1990s, the focus was mostly on transect studies that

provided basic information about gradients (Koch and Reddy 1992; Amador and Jones 1993;

Craft and Richardson 1993; Qualls and Richardson 1995; Amador and Jones 1995; Newman et


111

al. 1996 and 1998; Miao and Sklar 1998). Over the last decade, efforts were redirected to ex-

pand the spatial scope and sampling extension, leading to a more comprehensive sampling

scheme across the Everglades region, and a more integrated analysis (Noe et al. 2002; Noe et al.

2003; Daoust and Childers 2004; Chambers and Pederson 2006: Reddy et al. 2005) in some cas-

es, of changes to soil condition over time (Childers et al. 2003, Reddy et al. 2008). A brief de-

scription of some of the most important studies follows:

i. R-EMAP: This effort, initiated in 1993 by the USEPA Region 4, is one of the

largest and oldest spatially intensive monitoring programs for soils (along with

other compartments such as surface water, porewater, periphyton, macrophytes

and aquatic fauna) in the Everglades (Scheidt and Kalla 2007; Osborne et al.

2011a). Soil parameters measured include total phosphorus (TP), total nitrogen

(TN), total carbon (TC), methyl mercury, total mercury, organic matter content

and bulk density in surface soils and floc. This large scale project extended

across the EPA and the Big Cypress National Preserve, and utilized a probabil-

ity-based sampling design with a total of 1145 sites. The program has complet-

ed three phases from 1993-20051.

ii. ESM: The Everglades Soil Mapping Project was an effort conducted by Uni-

versity of Florida, in conjunction with the SFWMD and the RECOVER As-

sessment Team, during the period 2003-2004, covering the totality of the Ev-

erglades region (Reddy et al. 2005). A stratified random sampling method

was used to collect soil samples from 1,358 sites (plus 10 % replicates). Sam-

ples were collected at 3 depths: floc, surface (0–10 cm), and subsurface soils

(10–20 cm). Soil properties sampled included total phosphorus (TP), bulk

density (BD), moisture content (MC), loss-on-ignition (LOI), inorganic total

phosphorus (TPi), total nitrogen (TN), total carbon (TC), total calcium (TCa),

total magnesium (TMg), total iron (TFe), and total aluminum (TAl). The

ESM data has been used to: 1) investigate soil biogeochemical changes

through time and comparisons with previous spatial data (DeBusk et al. 2001;

1R-EMAP data available from USEPA at


112

Bruland et al. 2007; Grunwald et al. 2004; Grunwald et al. 2008; Marchant et

al. 2009); 2) explore innovative geostatistical techniques (including remote

sensing) to predict soil properties (Rivero et al. 2007a) and 3) to characterize

biogeochemical properties of specific areas (Bruland et al. 2006, Rivero et al.

2007a).2

iii. Florida International University: Four marsh transects were sampled by Doren

et al. in 1988-1989 (Doren et al. 1997) and then revisited by Childers et al.

(2003) in 1999-2000 to investigate the relationship between soil P and vegeta-

tion patterns. These transects covered Loxahatchee National Wildlife Refuge,

WCA-2A, 3A, and Shark River Slough, but Childers et al. (2003) added an-

other transect in Taylor Slough. Transects were anchored on major canals at

sources of water input. In the initial sampling they extended 5 to 8 km from

the canals, while in the Childers et al. (2003) sampling they extended 16 km.

Additional sampling has been collected by FIU. In January 2000, 49 soil pro-

files were examined (29 in the marsh and 10 in tree islands) in Shark Slough

(Ross et al. 2006a). The same research group collected and analyzed surface

soils from 464 locations in marl prairies and in 77 Everglades tree islands

(Ross and Sah in review).3

iv. Additional area specific sampling was conducted in the early 1990’s by the

Wetland Biogeochemistry Laboratory at the University of Florida with

SFWMD as a funding agency. This sampling event was conducted with a dif-

ferent approach than the transect studies and has been used for spatial model-

ing, based on a more robust sampling density and spatial distribution. These

studies have been reported as follows: LNWR (WCA-1) in 1991 (Newman et

2 ESM and historic sampling data available from SFWMD (contact person: Susan Newman) or Wetland Biogeochemistry Laboratory, University of Florida (contact person: Ramesh Reddy) 3 LTER data available from Florida International University available at http://fce.lternet.edu/data/ (contact person: Mike Rugge).


113

al. 1997), WCA-2A in 1992 (DeBusk et al. 1994) and WCA-3 in 1992 (Reddy

et al. 1993).

Available Models/Tools

The dynamics of peat soils in response to hydrology, nutrients, and fire seems to be well suited

to a modeling approach. Models that have been developed so far (Walker and Kadlec 1996;

Munson et al. 2002) have focused primarily on phosphorus storage in nutrient-enriched

compartments, and may not be applicable to the slow build-up (and occasional, sudden, fire-

induced loss) of soils in less impacted portions of the Everglades. A summary of these models is

shown in Table 2-3.

Peat Accrete (Larsen, Harvey et al. USGS)

Larsen et al. (2007) developed a conceptual model to simulate landscape biogeomorphology

patterning in the Everglades based on feedback between flow, sediment entrainment, and

deposition, and between nutrient delivery, vegetation distribution, and the differential

accumulation of peat (see Figure 2-9 from Larsen et al. 2007). PeatAccrete (currently version

1.0) is a numerical model of differential peat accretion feedback processes to test whether

observed landscape features and long-term stability can arise from this class of feedback

processes alone. This model serves as a first step toward validating the conceptual model as a

whole.

The PeatAccrete model utilizes two feedback mechanisms that govern landscape dynamics. The

first feedback mechanism involved differential peat accretion controlled by water levels and

phosphorus concentration, with the purpose of the “attainment” of an equilibrium ridge elevation

relative to slough. The second mechanism is characteristic of anabranching rivers, and it is a

feedback between channel morphology and sediment mass transfer controls lateral and

longitudinal topographic features. Anabranching rivers consist of parallel channels separated by

islands or ridges that are aligned with the flow and bed slope and remain stable on the timescale

of decades or longer (Nanson and Knighton 1996).


114

Everglades Phosphorus Gradient Model (EPGM)

Developed by William W. Walker and Robert H. Kadlec for the US Department of Interior in

1996 (Walker and Kadlec 1996, Kadlec and Walker 1999), the Everglades Phosphorus Gradient

Model (EPGM) is an expansion of the STA Design Model (STADM) that incorporates mass

balances on marsh water column and surface soils. The EPGM predicts variations in water

column P concentration, peat accretion rate, and soil P concentration along a horizontal gradient

imposed by an external P load and sheet-flow conditions (Figure 2-10).


115

Figure 2-10: Sequence of phosphorus mass balance models developed by Walker and Kadlec since the

initial STA design model developed in 1993. Figure adapted from Walker and Kadlec 2002.

Phosphorus Mass Balance Models

Phosphorus Mass Balance Models include the previously described model (EPGM) in addition to

two other models: STADM developed in 1993-1995, and the Dynamic Model for STA

(DMSTA), with several versions developed between 1999 and 2002. The integration of EPGM

in DMSTA (hybrid version) has been recently developed. These models, also created by Walker

and Kadlec in the time sequence shown in Figure 2-10, are based on mass balance principles, and

can be configured to work with existing hydrological models. DMSTA2 is designed to simulate

long-term responses in a marsh subject to a given P loading regime. While the initial model

(STADM) was developed to optimize the design of external P load controls, subsequent models

have been designed with the objective of simulating P impacts downstream of inflows, and

Evolution of Phosphorus Balance Models developed for Everglades

Water Column P

Water Column P

Water Column P

Water Column P

STA Design Model

1993-1995

EPGM1996-1997

Dynamic Model for STAs

1995-2002

EPGM/DMSTA Hybrid

Biomass P Biomass P

Soil PSoil P

Permanent Storage


116

optimization of restoration for water quality benefits. DMSTA and potentially the hybrid model

can be applied to soil types such as peat and marl.

ELM (Everglades Landscape Model)

This model was initiated in the early 1990s by University of Maryland’s Institute for Ecological

Economics under contract with SFWMD, and it has gone through several iterations (current

version ELM v.2.5) as well as an external peer review (Mitsch et al. 2007; Fitz and Trimble

2006). The ELM uses Spatial Modeling Environmental (SME) software developed by the

University of Maryland (Maxwell and Costanza 1995). The model is a regional scale simulation

model that integrates the hydrology, water quality, and biology of the various habitats in the

Everglades (Fitz and Trimble 2006). It has also been defined as a distributional mosaic model

that uses site-specific biogeochemical mechanisms and mass-balance to control energy and

material flows, and to predict changes in carbon and phosphorus structure of the soil, water, and

plant communities as a result of modified water deliveries to the Everglades (Sklar et al. 2001).

Geostatistical hybrid models with remote sensing

This model incorporates remote sensing into multivariate geostatistical models (regression

kriging and co-kriging) to predict P concentrations in floc and soil (Rivero et al. 2009). The

model has been tested and validated in WCA-2A, incorporating spectral data and derived indices

from two remote sensors (Landsat ETM+ and ASTER) with ancillary environmental data and

floc and soil TP measurements to predict the spatial distribution of floc and soil TP. Three spec-

tral indices derived from Landsat ETM+ and ASTER images: (a) Normalized Difference Vegeta-

tion Index (NDVI); (b) NDVI green; and (c) Normalized Difference Water Index (NDWI), and

environmental data: (a) distance to water control structures that serve as conduits for nutrient im-

port into WCA-2A; (b) distance to tree islands; (c) vegetation and land cover types; (d) x and y

geographic coordinates; and (e) principal component scores derived from Landsat ETM+ and

ASTER spectral datasets are evaluated in the model. Results from this model have been used to

quantify and provide visual representation of impacted and non-impacted areas, as related to re-

mote sensing indices as shown in Figure 2-11.


117

Figure 2-11: A representation of impacted and non-impacted areas in Water Conservation Area 2, based

on a vegetation index derived from Landsat ETM satellite imagery (NDVI green), used as a proxy to predict phosphorus concentrations in these area. Areas in gradations from dark blue to green are considered impacted, and areas in gradation of orange to yellow and con-sidered non-impacted (source: Rivero et al. 2009).


118

Table 2-3. Summary of existing soil models for the Everglades.

Model and

source

Category Purpose Input Output

Peat Accrete1 Numerical

model

Coupled morphological,

hydrological, and biologi-

cal models incorporating

feedback between vegeta-

tion, hydrology, and sedi-

ment transport to explain

vegetation patterning.

Dispersion and drag coefficients,

bed shear stress (as functions of

water-surface slope and depth-

averaged velocity); b) vegetation

community (community frontal

areas per unit volume, stem diame-

ters; c) expression for ridge and

slough drag coefficients derived

from dimensional analysis.

Simulated flows

and patterning of

longitudinal vege-

tational ridge and

slough.

ELM (Ever-

glades Land-

scape Model)2

Regional scale

simulation

model / Distri-

butional mosaic

model.

Dynamic integration of

hydrology, water quality,

and biology of Everglades

habitats to evaluate long-

term, regional benefits of

alternative project plans

with respect to water quali-

ty and other ecological

Performance Measures.

Current version: ELM v.

2.5

Water depth, soil nutrients, land

surface elevation, nutrient inflows,

periphyton and macrophyte bio-

mass. Boundary conditions include

rainfall, evapotranspiration, stage,

water flows through water control

structure and nutrients concentra-

tions at inflows.

P concentration

and net load along

spatial gradients

and other ecologi-

cal performance

measures, evaluat-

ed under scenarios

of alternative water

management plans.

Mass Budget

Models

(STADM,

EPGM,

DMSTA)3

Mass budget

model

Simulate P impacts down-

stream of inflows, and op-

timization of restoration

(reservoirs?) for water

quality benefits

Flow, water column and soil P, soil

accretion, biomass P, cattail densi-

ty.

Water and P mass

balance, water con-

centration profiles

and P accretion

rate.

Remote sens-

ing/ Geosta-

tistical Mod-

els4

Spatially ex-

plicit model

based on hybrid

geostatistical

model.

Prediction of soil and floc

TP concentration by inte-

grating remote sensing data

into multivariate geostatis-

tical models

Soil TP concentrations, vegetation

indices derived from satellite im-

agery (NDVI and NDVI green)

distance to canals, northing, vege-

tation at sampling locations.

Surface total P for

floc and soil at

various depths (0-

10 cm and 10-20

cm) 1 Larsen et al. 2007, Larsen and Harvey 2010 2 Fitz et al. 1995, Fitz and Trimble 2006 ELM Vers. 2.5 3 Walker and Kadlec 1996, Kadlec and Walker 1999 4 Rivero et al. 2007b, Rivero et al. 2009

2.0 Report of the Soils Subgroup · 2013-03-18 · Although always underfoot, soil is often a less...

Documents

Transcript of 2.0 Report of the Soils Subgroup · 2013-03-18 · Although always underfoot, soil is often a less...