This article appeared in a journal published by Elsevier. The... copy is furnished to the author for internal non-commercial research

This article appeared in a journal published by Elsevier. The... copy is furnished to the author for internal non-commercial research

This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
Author's personal copy
Ecological Engineering 37 (2011) 1481–1491
Contents lists available at ScienceDirect
Ecological Engineering
journal homepage: www.elsevier.com/locate/ecoleng
Legacy phosphorus in subtropical wetland soils: Influence of dairy, improved and
unimproved pasture land use
Ed.J. Dunne a,∗ , Mark W. Clark a , Ronald Corstanje b , K.R. Reddy a
a
b
Wetland Biogeochemistry Laboratory, Soil and Water Science Department, University of Florida/IFAS, 106 Newell Hall, P.O. Box 110510, Gainesville, FL 32611, United States
National Soils Resources Institute, Building 37, Cranfield University, Cranfield, Beds. MK4 OAL, UK
a r t i c l e
i n f o
Article history:
Received 24 September 2010
Received in revised form 28 March 2011
Accepted 10 April 2011
Available online 11 May 2011
Keywords:
Wetlands
Phosphorus
Soil
Land use
Legacy
a b s t r a c t
Wetlands provide various ecosystem services. One of these services includes nutrient storage in soils.
Soils retain and release nutrients such as phosphorus (P). This dynamic can be controlled by soil characteristics, overlying water quality, environmental conditions and historical nutrient loading. Historical
nutrient loading contributes to a legacy of P stored in soils and this may influence present day P dynamics
between soil and water. We quantified P characteristics of wetland soils and determined the availability
and capacity of soils to retain additional P loadings. We sampled surface (0–10) and subsurface (10–30)
wetland soils within dairy, improved and unimproved pastures. Surface soils had much greater concentrations of organic and inorganic P. Wetland soils in dairy had greatest concentrations of Ca and Mg,
probably due to inputs of inorganic fertilizer. They also had much greater total P, inorganic P, and P sorption capacity; however, these soils were P saturated and had little capacity to retain additional P loading.
Improved and unimproved pasture wetland soils had greatest amounts of organic P (>84%) and a capacity
to store additional P loadings. Using multivariate statistics, we determined that rather than being different based on land use, wetland soils in improved and unimproved pasture were dissimilar based upon
organic matter, organic P fractions, residual P, and soil metal (Fe and Al) content. The legacy of stored P in
soils, particularly wetland soils from dairies, combined with best management practices (BMPs) to reduce
nutrient loading to these systems, could contribute to a short-term release of soil-stored P to overlying
wetland water.
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
Wetlands provide many ecosystem services such as the storage
and transformation of nutrients and sediments lost from uplands
(Maynard et al., 2009; Noe and Hupp, 2009). In wetland ecosystems,
most nutrients, such as phosphorus (P), are stored in soils relative
to other ecosystem components such as vegetation, detritus, and
overlying water (Dunne et al., 2007; Dolan et al., 1981). Wetland
soils have an inherent ability to retain and release nutrients such as
P and this is dependent upon soil characteristics, overlying water
quality conditions and a range of environmental factors (Reddy
et al., 1999) that change through time and space. Soil characteristics
that are important for wetland soil P storage include soil organic
matter, aluminum (Al) and iron (Fe) content in acidic soils, and calcium (Ca) and magnesium (Mg) content in higher pH soils (D’Angelo
∗ Corresponding author. Present address: Division of Environmental Sciences, St.
Johns River Water Management District, 4049 Reid Street, Palatka, FL 32177, United
States. Tel.: +1 386 546 4529; fax: +1 352 392 3399.
E-mail address: [email protected] (Ed.J. Dunne).
0925-8574/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.ecoleng.2011.04.003
and Reddy, 1994; Braskerud, 2002; Bruland and Richardson, 2006;
Aldous et al., 2007; Bruland et al., 2009). Upon flooding, water
column properties such as pH, P and Ca concentrations play an
important role in P diffusion flux between overlying water and
underlying soil (Pant and Reddy, 2003). Other water-related environmental conditions that influence soil–water P dynamics include
water depth, duration of flooding and flooding frequency (Aldous
et al., 2005).
Implementing best management practices (BMPs) in agricultural landscapes have resulted in reduced P inputs into upland soils.
For instance, Sharpley (1999) estimated that P surpluses in agricultural systems reduced by 22 kg P ha−1 yr−1 . However, the positive
impact of BMPs can have varying successful effects on receiving
aquatic ecosystems (Hiscock et al., 2003; Kronvang et al., 2005;
Heckrath et al., 2008). It is often difficult to determine BMP effectiveness on receiving aquatic systems, as several factors influence
outcomes, with results often being site-specific (McCarty et al.,
2008; Merriman et al., 2009). A confounding factor can be the legacy
of historically stored nutrients in soils (Ekholm et al., 2005) due
to applying excessive amounts of fertilizers in uplands and subsequent loss and storage of nutrients in receiving wetland and aquatic
Author's personal copy
1482
Ed.J. Dunne et al. / Ecological Engineering 37 (2011) 1481–1491
ecosystems. In wetland and aquatic systems, the accumulated P
in soils and sediments can flux from soil to overlying water due
to concentration gradients between soils/sediments and the overlying water (Fisher and Reddy, 2001; Corstanje and Reddy, 2004;
Bostic and White, 2007). For example, Havens and James (2005)
indicate that when P loading is relatively low, shallow lakes (e.g.
Lake Okeechobee, FL) may effectively assimilate P, whereas after
many years of P loading, the assimilative capacity diminishes. Further, highly P-saturated lake sediments are often slow to respond to
reduced external loads because sedimentary stores of P can act as a
buffer to changes in water column P concentrations (Søndergaard
et al., 2005).
In the Lake Okeechobee Basin, FL, USA, four sub-basins, dominated by agriculture, have historically contributed most of the P
load to the lake (Flaig and Reddy, 1995). About 48% of the agricultural land is improved grazing pasture, 7% is dairy and 6% is in
unimproved pasture/rangeland (McKee, 2005). Within these four
sub-basins, 15% of the land area is wetland (Flaig and Havens,
1995) with over half being historically isolated from surrounding surface waters (but are now presently ditched) (McKee, 2005).
Isolated wetland systems can be defined as “wetlands completely
surrounded by uplands” (Tiner, 2003), which have occasional surface water connections to surrounding aquatic systems (Leibowitz,
2003; Leibowitz and Nadeau, 2003; Whigham and Jordan, 2003;
Winter and LaBaugh, 2003). Restoring the hydrology of these
presently ditched and drained wetlands may result in increased
amounts of water and P storage in the landscape, which could
help mitigate P loss from landscapes prior to waters reaching Lake
Okeechobee. However, cattle-grazed pasture and wetlands within
these pastures have probably received excessive P loads for many
years. Restoring and/or enhancing isolated wetlands to store more
P in soil may be confounded by the legacy of P accumulated in
soils. Therefore, it is important to determine the P characteristics
of wetland soils in these sub-basins and estimate the potential for
wetland soil to retain P and determine if wetland soils have any
additional P storage capacity. There is little information on isolated wetland soils at a sub-basin or basin wide spatial-scale in the
Lake Okeechobee Basin. However, a recent synthesis by Reddy et al.
(2011) did review soil P storage studies (a portion of which were
wetland studies) at the Greater Everglades Ecosystem scale, which
includes, but not limited to the Kissimmee River Basin, Okeechobee
Basin, Everglades Agricultural Areas, Stormwater Treatment Areas
and Everglades National Park, as it relates to restoration activities
in South Florida. Other studies report soil P characteristics of specific wetland sites (Bridgham et al., 2001; Aye et al., 2006; Dunne
et al., 2006b; Bostic and White, 2007).
It is critical for land and water managers to know P characteristics and the potential for P retention and release by soils given the
uncertainty surrounding the potential positive BMP impacts unto
improving water quality.
The objectives of our study were to (1) quantify the soil physicochemical characteristics and P characteristics of isolated wetland
soils from dairy, improved and unimproved pastures within subbasins of the Okeechobee Basin, (2) determine the labile and
non-labile P fractions in soils, (3) quantify soil P availability and
capacity of soils to retain additional amounts of P using indices, and
(4) use multivariate statistics to investigate relationships between
land use, soil physicochemical and P characteristics along with the
various P indices that we used.
ing pasture (Fig. 1). Pastures within dairies were cattle grazed areas
adjacent to dairies. Both dairy and improved pastures were dominated by Paspalum notatum Flugge (Bahia grass). Improved pastures
were fertilized with nitrogen and grazed by cattle primarily in the
wet season, which is from May to October. Unimproved pastures
were areas dominated by a mixture of both Bahia grass and native
grasses. These pastures are typically not fertilized and are often
grazed in the dry season, which is between about November and
April (Gathumbi et al., 2005; Dunne et al., 2007).
Sites were identified using data from a previous synoptic survey
of 118 wetlands within the four sub-basins known as the “four
priority basins” of the Okeechobee Basin (McKee, 2005). The four
priority basins are north of Lake Okeechobee and are 121,000 ha in
area, with 64% being agriculture (SFWMD, 2003). Wetlands were
selected based on agricultural land use, soil total P concentrations,
soil organic matter, and soil metal content that were previously
quantified during the earlier synoptic survey.
2.2. Sample collection
Wetland sites were sampled during October and November
2005. We collected six soil samples from each wetland. We adopted
a stratified random sampling approach. Three soil samples were
collected from the center or deep marsh zone of the wetland. Three
other samples were taken from the surrounding wetland edge or
“shallow marsh” zones (Dunne et al., 2007). Soils were sampled
to a depth of 30 cm using a simple coring device. A polycarbonate tube (7.5 cm diameter × 30 cm length) was sharpened at one
end and hammered down to a depth just below 30 cm. Tubes with
soil in them were extracted from soil and extruded at depth specific increments. Depth increments were typically 0–10 cm and
10–30 cm; however, this varied, depending on soil horizon development. In some instances, three depth increment samples were
collected per soil core. Extruded depth specific samples were placed
into pre-labeled zip lock bags and put on ice in a cooler. Soils were
transported back to the Wetland Biogeochemistry Laboratory, Soil
and Water Science Department at the University of Florida, for
preparation and analyses.
2.3. Water sample collection and laboratory analyses
All wetlands were not sampled for overlying water, as not all
sites were flooded during sampling. When wetland sites were
flooded to about 30 cm, we sampled site waters. Site waters were
sampled and analyzed for soluble reactive P (SRP), total dissolved
P (TDP), and total P (TP). Water samples were filtered to 0.45 ␮m in
the field and later analyzed in the laboratory for SRP using Method
365.1 (USEPA, 1993). Total dissolved P was measured on a filtered
water sample (0.45 ␮m) that was digested with potassium persulfate. The digestate was then analyzed for P using an automated
ascorbic acid method (Method 365.1; USEPA, 1993). Total P was
determined on unfiltered sample that was digested and analyzed
for P as previously described.
During site sampling, we also recorded site water physical
chemistry using a YSI 556 multiparameter system. Wetland water
temperature, specific conductance, dissolved oxygen content and
oxygen reduction potential were measured.
2.4. Soil physicochemical and phosphorus characteristics
2. Methods
2.1. Site selection
We sampled 20 isolated wetlands within dairy pasture,
improved cow-calf grazing pasture, and unimproved cow-calf graz-
All soil samples were measured for: pH; water content;
bulk density; organic matter; TP; inorganic phosphorus (TPi )
extracted with 1 M HCl; total carbon (TC) and total nitrogen (TN);
water extractable P (WEP); acid extractable P (PHCl ), Ca (CaHCl ),
Author's personal copy
Ed.J. Dunne et al. / Ecological Engineering 37 (2011) 1481–1491
1483
Fig. 1. Outline of Florida and main watersheds. Location of wetland sample sites in dairy, improved and unimproved land uses. Sub-basin boundaries are outlined.
Mg (MgHCl ), Al (AlHCl ) and Fe (FeHCl ) and; ammonium oxalate
extractable P (Pox ), Fe (Feox ), and Al (Alox ).
Soil pH was measured in a 1:2 soil to water ratio. A known mass
of wet soil was dried for 72 h at 70 ◦ C and the percentage difference
was quantified as soil water content. Soil bulk density was determined on a dry weight basis, with a known volume of soil using a
coring method. Soil total P content was determined on 0.5 g of finely
ground dry soil combusted at 550 ◦ C in a furnace for 4 h. Ash was
then dissolved in 6 M HCl (Andersen, 1976) and digestate analyzed
for P using an automated ascorbic acid method (Method 365.1;
USEPA, 1993). Dried, finely ground soil was analyzed for TC and
TN by dry combustion using a C–N–S analyzer (Carlo Erba Model
NA-1500). To determine inorganic P, soils were dried, ground and
sieved to 2 mm. Soils were then extracted with 1 M HCl. Soil solutions were centrifuged and filtered to 0.45 ␮m using a vacuum
filtration system. Soluble reactive P was analyzed using an automated ascorbic acid method as previously mentioned. To quantify
WEP, field moist soils were extracted with dissolved deionised
water for 1 h. Samples were then centrifuged, filtered to 0.45 ␮m,
and analyzed for SRP as previously described. Metal content (CaHCl ,
MgHCl , FeHCl and AlHCl ) was also analyzed on soils extracted in
1 M HCl using inductively coupled plasma-atomic emission spectroscopy (Thermo Jarell Ash ICAP 61E, Franklin, MA). To determine
ammonium oxalate extractable Pox , Feox and Alox , air-dry soil samples were extracted (1:50 soil/sediment to solution ratio) with
0.2 M oxalic acid + 0.175 M ammonium oxalate (adjusted to pH 3.5)
(McKeague and Day, 1966) and filtered solutions were analyzed for
P, Fe and Al using atomic absorption spectroscopy.
2.5. Soil phosphorus fractionation
Soil P was fractionated using a similar scheme to that described
by Ivanoff et al. (1998). Phosphorus fractions were inorganic P
extracted with NaHCO3 (Al/Fe bound P), inorganic P extracted
with HCl (Ca/Mg bound P), bioavailable organic P (BOP), microbial
biomass P (MBP), fulvic acid bound P (FAP), humic acid bound P
(HAP), and residual non-reactive P (ResP). Soil P fractionation was
undertaken on surface and subsurface soils collected from four of
dairy, improved and unimproved pasture wetland sites. In total,
about 55 soil samples were analyzed.
Total inorganic P was the sum of inorganic P in a non-fumigated
soil extracted with 0.5 M NaHCO3 and the inorganic P in a 1 M
HCl extract; BOP was the total P in a non-fumigated soil extracted
with 0.5 M NaHCO3 minus the inorganic P in a non-fumigated soil
extracted with 0.5 M NaHCO3 ; MBP was the TP in a fumigated 0.5 M
NaHCO3 extract minus the TP in the non-fumigated extract; FAP
was the TP in pretreated 0.5 M NaOH extract; whereas HAP was
the TP in a 0.5 M NaOH extract minus the TP in pretreated a 0.5 M
NaOH extract; and finally, ResP was the TP in remaining soil residue
after all previous extractions.
2.6. Phosphorus sorption indices
A phosphorus sorption capacity index (PSI) was determined on a
moist soil incubated at 1000 mg P kg−1 for 24 h (Reddy et al., 1998).
Soil extracts were then centrifuged, filtered and filtrate was analyzed for SRP as previously described. The amount of P sorbed by
soil was the difference between the initial and final concentration
of SRP in the soil-extracted solution, expressed on a soil dry weight
basis.
The P saturation ratio (PSR) was calculated as:
PSR =
POX
FeOX + AlOX
where Pox , Feox , Alox are expressed in molar mass (mmol kg−1 ).
Author's personal copy
1484
Ed.J. Dunne et al. / Ecological Engineering 37 (2011) 1481–1491
Table 1
Physicochemical characteristics of surface and subsurface wetland soils collected from dairy, improved and unimproved pasture. Soils were sampled during October and
November 2005. Surface soils were typically sampled to a depth of 10 cm, while subsurface soils were typically sampled between 10 and 30 cm depth. Within subsurface
depths, soils were often sectioned into additional depth specific samples, which were based upon horizon breaks between 20 and 30 cm. Values in the table are means ± one
standard error.
Land use
Depth
n
pH
Bulk density (g cm−3 )
Water content (%)
Organic matter (%)
Total nitrogen (%)
Total carbon (%)
Dairy
Surface
Subsurface
24
27
6.1 ± 0.2
5.9 ± 0.2
1.00 ± 0.1
1.25 ± 0.1
56 ± 4
38 ± 5
38 ± 6
27 ± 6
1.2 ± 0.2
0.8 ± 0.2
19.0 ± 2.9
14.9 ± 3.6
Improved
Surface
Subsurface
68
90
5.1 ± 0.1
5.0 ± 0.1
1.01 ± 0.0
1.35 ± 0.0
46 ± 2
29 ± 2
22 ± 2
12 ± 2
0.7 ± 0.1
0.3 ± 0.1
10.6 ± 1.1
5.6 ± 0.9
Unimproved
Surface
Subsurface
24
26
4.3 ± 0.1
4.4 ± 0.1
0.78 ± 0.1
1.32 ± 0.1
44 ± 4
23 ± 3
30 ± 4
11 ± 3
0.9 ± 0.1
0.3 ± 0.1
15.1 ± 2.2
5.6 ± 1.7
Soil P storage capacity (SPSC) is an indicator of a soil’s ability
to sorb additional P before exceeding a threshold concentration, at
which point P levels are of environmental concern (Nair and Harris,
2004). Nair and Harris (2004) estimated SPSC as:
dairy pastures (p < 0.05). These one-time sampling results suggest
that dairy impacted wetland waters have about 10 times the P relative to improved pasture wetlands and about 60 times the P relative
to unimproved pasture wetland waters.
SPSC = (0.15 − PSR) × (AlOX + FeOX )
where 0.15 is a critical PSR that contributes to eutrophication
(Nair et al., 2004). Chrysostome et al. (2007) suggested that Nair
et al. (2004) chose a 0.15 PSR, as this was a threshold for Florida
upland agricultural soils to approximate a threshold concentration
of 0.10 mg P L−1 in solution. We calculated P sorption indices for all
surface and subsurface soils.
2.7. Statistical analyses
Data distributions for soil physicochemical, P characteristics and
P indices were tested for normality. If data were non-normal prior
to statistical analyses, they were log transformed to approximate
normality. Parametric statistical analyses were conducted on logtransformed data and significant differences between groups were
determined at the p < 0.05, 0.01, and 0.001 level using t-tests and
analysis of variance (ANOVA).
If data did not approximate normality after transformation, then non-parametric tests such as the Mann–Whitney and
Kruskal–Wallis were undertaken to test for ranked differences. For
example, non-parametric tests were undertaken on P indices data.
Multivariate analyses were also undertaken and tests included
principal component analysis (PCA), cluster analysis and discriminant analysis. These analyses were conducted on log-transformed
data. Multivariate analyses were all undertaken using the Statistica Software package Version 9.1, Oklahoma, USA, while univariate
statistics were calculated using Minitab for Windows Version 15,
Pennsylvania, USA.
3. Results and discussion
3.1. Water
At the time of sampling (October and November 2005)
dairy wetland waters tended to have greater temperatures
(25 ± 1.7 ◦ C; mean ± 1 standard deviation), greater specific conductance (1.35 ± 1.2 mS cm−1 ) and greater pH (7.2 ± 0.5) than wetlands
located within improved and unimproved pasture. When open
water was present in wetlands, P concentrations were greatest in
dairy pasture wetlands (2.5 ± 0.83 mg TP L−1 ), which were greater
(p < 0.05) than concentrations in improved pasture wetland waters
(0.29 ± 0.05 mg TP L−1 ), which in turn, were greater (p < 0.05)
than P concentrations in unimproved pasture wetland waters
(0.04 ± 0.001 mg TP L−1 ). We observed similar patterns in TDP and
SRP. Greatest concentrations of TDP (1.92 ± 0.8 mg L−1 ) and SRP
(1.96 ± 0.7 mg L−1 ) were in overlying wetland waters located in
3.2. Soil physicochemical and phosphorus characteristics
Wetland soils in dairy pasture tended to have greater pH, Ca,
and Mg concentrations than soils collected from improved and
unimproved pasture wetlands (Tables 1 and 2; p < 0.05). This was
probably due to increased inputs of food and fertilizer in dairy
pasture relative to the other land uses. For example, Hiscock et
al. (2003) reported that net P import coefficients for dairies in
the Okeechobee Basin were 54 kg ha−1 yr−1 , whereas imports for
improved and unimproved pasture were about 3 and <0.05 kg
P ha−1 yr−1 , respectively.
Wetland soils in improved and unimproved pasture had similar
soil Mg and Ca concentration; however, wetland soils in improved
pasture had greater soil pH than unimproved pasture wetland soils
(p < 0.05). Dairy wetland soils also had the greatest soil organic
matter content (Table 1; p < 0.05). This was surprising, as during
sampling, most dairy wetlands were devoid of emergent vegetation, which is the precursor for detritus that contributes to
accreting soil organic matter. We hypothesize that rather than a
veneer of surface organic material, the organic matter was mixed
with overlying waters, surface soils and subsurface soil layers, due
to cattle disturbance. During sampling, we visually observed a lot
of organic material suspended in wetland water columns and soil
surfaces were poached by cattle.
Surface soils had greatest concentrations of most soil characteristics measured. These include water content, organic matter, TP,
WEP, PSI, TN, TC, CaHCl , MgHCl , and Fe (Tables 1–3; p < 0.05). Further, surface soils had much lower bulk density than underlying
subsurface soils (Table 1; p < 0.001), which was probably due to its
slightly higher organic matter and soil water content. Surface and
subsurface soils had similar soil pH (Table 1) and Al content was
distributed relatively uniform throughout the 30 cm soil depth that
was sampled (Table 2).
Total P concentration in dairy soil (both surface and subsurface; 0–10 cm and 10–30 cm) was nearly three times greater than
the concentrations in improved and unimproved pasture wetland soils (Table 3; p < 0.001) suggesting that dairy wetland soils
were P enriched. Dairy wetland soils also had greatest concentrations of soil TPi and WEP (Table 3; p < 0.001). Wetland soils in
improved pasture had similar TP concentrations to unimproved
pasture wetland soils with concentrations being medium to low
(<600 mg P kg−1 ). DeBusk et al. (2001) suggest that P concentrations in wetland soils greater than 500 mg P kg−1 be considered
“P-enriched.” A later study on surface (0–10 cm) wetland soils
collected from various types of wetlands in the southeastern US
suggest that background soil TP concentrations of least impacted
Author's personal copy
Ed.J. Dunne et al. / Ecological Engineering 37 (2011) 1481–1491
1485
Table 2
Surface and subsurface soil metal characteristics collected from wetlands within dairy, improved and unimproved pasture. Surface soils were typically sampled to a depth of
10 cm, while subsurface soils were typically sampled to a depth of 30 cm. Within subsurface depths, soils were often sectioned into additional depth specific samples, which
were based upon horizon breaks within the 10–30 cm depth increment. Soils were sampled during October and November 2005. Values are means ± one standard error.
Land use
Depth
1 M HCl
0.2 M ammonium oxalate
−1
−1
n
Ca (mg kg
)
Mg (mg kg
−1
)
Fe (mg kg
−1
)
Al (mg kg
)
Fe (mg kg−1 )
Al (mg kg−1 )
Dairy
Surface
Subsurface
24
27
4715 ± 1036
1141 ± 216
828 ± 161
232 ± 47
296 ± 48
130 ± 31
4027 ± 978
3947 ± 1631
609 ± 88
607 ± 179
4184 ± 1088
4042 ± 1450
Improved
Surface
Subsurface
68
90
1741 ± 174
644 ± 106
248 ± 21
83 ± 17
473 ± 45
293 ± 53
2272 ± 537
1914 ± 378
663 ± 62
431 ± 59
1798 ± 392
1576 ± 282
Unimproved
Surface
Subsurface
24
26
1189 ± 225
403 ± 123
277 ± 49
142 ± 55
468 ± 57
176 ± 30
1831 ± 494
1764 ± 526
707 ± 91
230 ± 41
2196 ± 596
1748 ± 521
Table 3
Phosphorus characteristics of surface (0–10 cm) and subsurface (10–30 cm) wetland soils collected from wetlands located in dairy, improved and unimproved pastures.
Values are means ± one standard error. Soils were sampled during October and November 2005.
Parameter
Dairy
Total P (TP), mg kg−1
Inorganic P (TPi ; 1 M HCl), mg kg−1
P sorption capacity (PSI), mg kg−1
Water extractable P (WEP), mg kg−1
Ammonium oxalate P (Pox ), mg kg−1
Soil phosphorus storage capacity (SPSC), mmol kg−1
1253
717
512
22
1094
−11
Improved
Surface
±
±
±
±
±
±
Subsurface
230
181
70
5
214
7
wetland sites sampled were about 550 mg kg−1 (Mukherjee et al.,
2009).
Depending on land use, inorganic P was 60%, 14%, and 12% (dairy,
improved, unimproved, respectively) of soil total P storage (g m−2 )
(Fig. 2; Table 3). Water extractable P concentrations, which is P
that is directly available to low P waters (soils were extracted with
DDI water that contains little P, <0.01 mg P L−1 ) was a small portion of soil total P (Table 3). Dairy soils contained about 2.4% as
WEP, whereas improved and unimproved pasture wetland soils
contained less than 1.5% of soil TP as WEP.
Many soil physicochemical characteristics were linearly related
and significant at the p < 0.001 level. Some of the best relationships were between soil water content and soil organic matter
(r = 0.92), total N and total C (r = 0.98); total inorganic P and total
P (r = 0.92). These significant correlations are a likely result of
many soil physicochemical and P characteristics containing redundant and overlapping information. For example, TP, inorganic P
extracted with 1 M HCl and WEP. The use of principle component analysis (PCA) is an effective way to reduce this redundancy
(Savvides et al., 2010) as it transforms correlated data, in our case,
soil characteristics, into a small number of uncorrelated principle
components. Using PCA, we found that four principal components
explained about 90% of the variation in soil physicochemical and P
characteristic data. The first two principle components described
the majority (70%) of the variation (Table 4). Fig. 3 is a biplot of
the first two principal components, in which wetland soils within
the different land uses were identified. Dairy pasture wetland soils
were distinct from unimproved and improved pasture wetland
soils, with little difference between the latter two. The separation between dairy pasture wetland soils and the other two was
mainly across principle component 2, which only accounts for 22%
of the variation. This suggests that other factors, independent of
land use, contribute to varying soil P in improved and unimproved
pasture wetland soils. Soil characteristics with largest eigenvalues
for principle component 2 were Fe, bulk density, TP, Pox , TPi , and
pH (Table 4). Soil water content, TN, TC and soil organic matter
were important eigenvectors for principle component. Soil characteristics of component 2 suggest that these soils are mineral-like,
394
245
402
8
356
14
±
±
±
±
±
±
67
80
68
2
77
8
Surface
354
44
271
1
187
7
±
±
±
±
±
±
34
4
32
0
24
2
Unimproved
Subsurface
147
18
222
1
102
12
±
±
±
±
±
±
Surface
18
2
29
0
15
2
315
37
303
2
129
10
±
±
±
±
±
±
33
4
68
1
22
3
Subsurface
107
10
209
1
72
13
±
±
±
±
±
±
23
2
52
0
19
3
Table 4
Eigenvectors associated with the first two principal components for soil physicochemical and phosphorus (P) characteristics. Dataset includes surface and
subsurface soils collected from 20 wetlands from three different land uses in Okeechobee Basin, FL. Soil metal content denoted with ox or HCl (in subscripts) are those
metals and P extracted with 0.2 M oxalic acid + 0.175 M ammonium oxalate or 1 M
HCl.
Eigenvectors
PC 1
PC 2
AlHCl
Alox
Bulk density
CaHCl
FeHCl
Feox
MgHCl
Organic matter
pH
Pox
Total C (TC)
Total inorganic P (TPi )
Total N (TN)
Total P (TP)
Water content
0.25
0.25
−0.23
0.22
0.23
0.23
0.26
0.32
−0.02
0.26
0.33
0.15
0.33
0.24
0.30
0.11
0.11
0.25
0.07
−0.27
−0.27
0.09
−0.12
0.42
0.30
−0.12
0.42
−0.13
0.30
−0.14
whereas soil characteristics in principle component 1 are diagnostic of organic like soils.
We subsequently undertook cluster analysis on soil physicochemical and P characteristics data to establish if the observed data
structure in Fig. 3 represents underlying data groups. We found
three clusters that showed membership as a function of the first
two principal components (Fig. 4). One of the clusters matches
dairy pasture wetland soils (Cluster 2). However, improved and
unimproved pasture wetland soils separate out into two different clusters that seem independent of land use and are primarily
a function of the variability described by principle component
1. Groupings or clusters of observations identified in a principal component analysis do not always reflect the underlying data
structure. For example, we observed a set of groups that the PCA
did not resolve; therefore, we undertook cluster analysis. When
groupings of observations are observed in principle component
Author's personal copy
1486
Ed.J. Dunne et al. / Ecological Engineering 37 (2011) 1481–1491
Surface soils
Subsurface soils
100
Total P g m -2
Total P g m
-2
100
10
1
1
0.1
0.1
Dairy
Improved Unimproved
100
100
10
10
1 M HCl P g m -2
1 M HCl P g m -2
10
1
Dairy
Improved Unimproved
Dairy
Improved Unimproved
1
0.1
0.1
0.01
0.01
Dairy
Improved Unimproved
Fig. 2. Box plots of mass areal storage of total phosphorus and inorganic phosphorus as extracted with 1 M HCl in surface and subsurface wetland soils collected from dairy,
improved and unimproved pastures.
Fig. 3. Biplot of principle components for soil physicochemical and phosphorus
characteristics of dairy, improved pasture and unimproved pasture wetland soils.
Fig. 4. Cluster analysis of the principle components for soil physicochemical and
phosphorus characteristics of dairy, improved pasture and unimproved pasture
wetland soils.
Author's personal copy
Ed.J. Dunne et al. / Ecological Engineering 37 (2011) 1481–1491
1487
Table 5
Results from the stepwise discriminant analysis of surface and subsurface soils collected from 20 wetlands sampled in the Okeechobee Basin,
FL. Soil characteristics in the first column are those selected by the stepwise procedure, the second and third column contain the coefficients of
the linear discriminant functions (DF) and the fourth column, is the assigned cluster. Within model 1, the soil metal and phosphorus (P) content
denoted with ox or HCl (in subscripts) are those metals and P extracted with oxalic acid and ammonium oxalate or HCl. Water extractable
P = WEP; total carbon, nitrogen, and phosphorus = TC, TN, and TP; the P extracted with HCl = TPi . Model 2 includes P in non-fumigated soils
extracted with NaHCO3 (NaHCO3 Pi -NF); HCl-Pi = P in soils extracted with HCl; BOP = biologically available organic P; MBP = microbial biomass
P; FAP = fulvic acid bound P; HAP = humic acid bound P; and ResP = residual P. Model 3 includes PSI, which is the P sorption capacity index
(g m−2 ); SPSC = soil P storage capacity (g m−2 ); and PSR = P saturation ratio (PSR).
Characteristic
DF1
Model 1: Soil physicochemical and phosphorus characteristics
2.56
AlHCl
Alox
−2.68
CaHCl
−0.35
WEP
0.47
FeHCl
−2.11
Feox
3.04
0.072
MgHCl
1.05
Organic matter
−0.19
pH
−0.35
Pox
TC
−1.16
TN
0.42
TP
−0.08
−1.05
TPi
Model 2: Soil phosphorus fractionation
−0.06
BOP
FAP
1.27
HAP
0.64
−2.03
HCl-Pi
MBP
0.001
−1.23
NaHCO3 Pi -NF
0.74
ResP
Model 3: Phosphorus sorption indices
−0.88
PSI
−1.75
PSR
SPSC
−0.00053
a
DF2
Characterizesa
−3.50
2.05
−0.21
−0.33
2.13
−2.25
0.48
0.76
0.012
1.52
−0.44
0.54
−0.46
−0.13
Cluster 3
Cluster 2
1.58
−1.63
−1.00
0.51
−0.21
0.17
0.68
Cluster 3
Cluster 1
Cluster 1
Cluster 2
0.21
0.166
0.0010
Cluster 2
Cluster 3
Cluster 1
Clusters 1 and 3
Cluster 2
Cluster 2
Cluster 2
Cluster 3
Cluster 2
Cluster 2
Where no cluster was identified, that soil characteristic was not important in differentiating between clusters.
analysis, a formal ‘post-hoc’ test is implemented using an ANOVA.
We extend this using discriminant analysis, a multivariate form
of ANOVA. This analysis procedure selects and combines predictor
variables in linear combinations to discriminate between groups
(Corstanje et al., 2009; McCune and Grace, 2002). The objective
of our discriminant analysis was to determine; (i) how the soil
physio-chemistry determines the soil characteristics found in the
different land uses, (ii) whether P fractionation support our findings
on the soil physio-chemistry, and (iii) if P indices are an effective
approach to describing these particular soil characteristics for each
land use. This cannot be determined using principle component
analysis alone. In Table 5, we present the discriminant functions
developed on soil physicochemical and P characteristics as Model
1. In this table, the first discriminant function (DF) separated Cluster 2 (dairy; negative scores for DF1) from the two other clusters
(positive scores for DF1). The second function separated Cluster 1
(positive scores for discriminant function 2) from Cluster 3 (negative scores for DF2). The discriminant function 1 coefficients for TPi ,
Alox and FeHCl are large and negative; therefore, these soil characteristics contribute to discriminating Cluster 2 from Clusters 1 and
3. Dairy pasture soils in Cluster 2 are distinct in mostly mineral
like soil characteristics; however, they are also quite different in
organic characteristics such as soil TC (Table 5). When we considered the combined Clusters 1 and 3, and compared this to Cluster
2 (dairy), we found that improved and unimproved wetland soils
have distinctly different inorganic and organic P content to dairy
wetland soils.
The second discriminant function in Table 5 identifies the differences between soil groupings within improved and unimproved
pasture wetland soils. Rather than being distinct based on land use,
these soils were different based upon soil organic matter (Clus-
ter 1), whereas Cluster 3 had distinct concentrations of Feox and
AlHCl . These results show that dairy land management has a significant impact on soil properties, with larger amounts of inorganic
P associated with Alox and FeHCl . Wetland soils within improved
and unimproved pastures were characterized predominantly by
whether soils were mineral or organic.
3.3. Soil phosphorus fractionation
When we compared soil P fractions with soil sampling depth,
we found that surface soils had significantly greater concentrations
of Al/Fe bound P (twice the amount; 2×), Ca/Mg bound P (2×), BOP
(2×), MBP (7×), FAP (8×), HAP (2×) and ResP (3×) than underlying
soils to a maximum depth of 30 cm (p < 0.05). However, when we
expressed these concentrations as a percentage of soil TP, patterns
changed. For example, subsurface soils had a significantly greater
percentage of soil total P stored as Fe/Al bound P and BOP (p < 0.05),
while percentages of Ca/Mg bound P, FAP, HAP, and ResP did not
change with depth (Fig. 5). Surface soils had a greater proportion of
soil TP stored as MBP (p < 0.05). Pooling the various P fractions into
larger groupings (inorganic and organic P fractions) also suggested
that soils collected from the different land uses had different soil P
availabilities. For example, wetland soils from dairy pastures stored
62% of soil P in inorganic fractions (Al/Fe bound P + Ca/Mg bound
P), which was significantly greater than the inorganic P fractions
in improved (15%) and unimproved (11%) pasture wetland soils
(Fig. 5; p < 0.05). Improved and unimproved pasture wetland soils
contained about twice (22% of soil total P) the amount of available organic P (BOP + MBP) relative to dairy pasture wetland soils.
Organic P fractions must undergo transformation to inorganic P
forms prior to becoming bioavailable to the overlying water col-
Author's personal copy
1488
Ed.J. Dunne et al. / Ecological Engineering 37 (2011) 1481–1491
Fig. 5. Soil phosphorus fractions in surface (0–10 cm) and subsurface (10–30) wetland soils from dairy, improved, and unimproved pasture. All fractions are expressed as a
percent of soil total phosphorus concentration (mg kg−1 ). NaHCO3 Pi -NF = soluble reactive P in non-fumigated soils extracted with NaHCO3 ; HCl-Pi = soluble reactive P in soils
extracted with HCl; BOP = total P in non-fumigated soil extracted with NaHCO3 − soluble reactive P in non-fumigated soils extracted with NaHCO3 ; MBP = microbial biomass
P (total phosphorus in non-fumigated NaHCO3 soil extract − total P in fumigated NaHCO3 soil extract), FAP = fulvic acid bound P (total P in pretreated NaOH soil extract);
HAP = humic acid bound P (total P in NaOH soil extract − total P in pretreated NaOH soil extract); and ResP = residual P.
umn (Reddy et al., 2005). Slowly available organic P fractions such
as FAP and HAP (which is P associated with decomposed biological
material) were both a greater proportion of soil total P in improved
and unimproved pasture wetland soils, relative to dairy pasture
wetland soils (Fig. 5; p < 0.01), while improved and unimproved
soils had similar amounts. Dairy wetland soils stored about 38% of
soil P in organic P fractions, whereas improved and unimproved
soils stored about twice that amount, 85 and 89%, respectively.
Cheesman et al. (2010) suggested that similar isolated wetland soils
collected from improved pastures contained about 73% of total P in
organic forms, as determined using solution 31 P nuclear magnetic
resonance. The majority of this pool composed of phosphomonoesters. In our sampled soils, less than 16% of total P was stored
as ResP. Residual P fractions are considered to represent resistant
organic P and/or unavailable mineral P (Reddy et al., 2005; Turner
et al., 2005).
We found linear relationships between total inorganic P and
Fe/Al bound P (r = 0.89), and Ca/Mg bound P (r = 0.87); total P and
Ca/Mg bound P (r = 0.81); WEP and Al/Fe bound P (r = 0.87); Ca and
Mg were related to ResP (r = 0.65); Alox content with FAP (r = 0.72)
and HAP (r = 0.73); and BOP with FAP (r = 0.82).
In Table 5, Model 2 considers whether the more detailed P
fractionation scheme supports our findings from the soil physicochemical characteristics. Similar to what we observed for Model
1, this analysis showed that Cluster 2 (dairy pasture wetland soils)
is distinct from Cluster 1 and Cluster 3 based on inorganic P fractions (NaHCO3 Pi -NF and HCl-Pi ). Soils in Cluster 1 had increased
amounts of slowly available organic P fractions (FAP and HAP)
(Reddy et al., 1999). In this model, Cluster 3 soils were characterized by ResP and BOP, implying that ResP fractions represented
unavailable mineral bound P, rather than highly resistant organic P
fractions. Wetland soils are often characterized as being anaerobic,
having large amounts of organic matter and storing a large proportion of soil total P in organic P forms. Although decomposition is
slow under anaerobic conditions, the mineralization of organic P
to inorganic available forms plays an important role in long-term
soil P storage (Reddy et al., 1999).
3.4. Phosphorus sorption indices
Dairy pasture wetland soils had the greatest P sorption capacity relative to improved and unimproved pasture wetland soils
(Table 3; Fig. 6; p < 0.01). Dairy soils had about twice the P sorption capacity (median = 31.6 g m−2 ) relative to improved pasture
soils, with improved pasture soils (14.1 g m−2 ) having about twice
the amount as unimproved. This suggests that although dairy wetland soils contained large amounts of available P, they also had
capacity to sorb P. Other studies suggest that as P loading increases,
P retention can also increase (Reddy et al., 1999; Dunne et al.,
2006a). Increased concentrations of Ca and Mg in dairy wetland
soils may contribute to increased P sorption capacity (Table 3). The
Fe content of wetland soils was similar between land use types
(Table 2); however, the Al content was greatest in dairy wetland
soils (p < 0.001). This could contribute to these soils having greater
P sorption relative to other wetland soils, as Al content is important
for P dynamics, with Al bound P being affected by changes in pH
rather the changes in soil redox (D’Angelo, 2005).
Dairy wetland soils however had much greater PSR values, suggesting that these soils (both surface and subsurface) were more
P saturated relative to improved and unimproved pasture wetland
(Fig. 6).
When we investigated whether wetland soils had additional
capacities to store P (SPSC), we found that dairy soils were indeed
P saturated (Table 3; Fig. 6) and had no capacity to store additional
amounts of P. Surface wetland soils from both improved and unimproved pasture had similar abilities to retain additional amounts of
P. Subsurface soils from all land uses had similarly low SPSC values.
The P indices considered in this study can also be used as predictors of the three cluster groups observed. Model 3 as shown in
Table 5 indicated that dairy wetland soils (Cluster 2) have distinctly
Author's personal copy
Ed.J. Dunne et al. / Ecological Engineering 37 (2011) 1481–1491
PSI g m-2
Surface Soils
Subsurface Soils
140
140
120
120
100
100
80
80
60
60
40
40
20
20
0
0
-20
-20
Dairy
Improved Unimproved
Dairy
Surface Soils
SPSC g m-2
1489
Subsurface Soils
300
300
200
200
100
100
0
0
-100
-100
-200
Improved Unimproved
-200
Dairy
Improved Unimproved
Dairy
Surface Soils
Improved Unimproved
Subsurface Soils
10
1
1
0.1
0.1
0.01
0.01
0.001
0.001
PSR
10
0.0001
0.0001
Dairy
Improved Unimproved
Dairy
Improved Unimproved
Fig. 6. Box plots of phosphorus indices that are the phosphorus sorption capacity index (PSI), soil phosphorus storage capacity (SPSC) (both expressed on a storage basis;
g m−2 ), and phosphorus saturation ratio (PSR) in surface and subsurface wetland soils collected from dairy, improved and unimproved pasture.
different PSI and PSR values and were separated from other clusters along DF1. However, the indices were not able to discriminate
between the groups observed in improved and unimproved soils.
Therefore, the P indices used are effective indicators of P enrichment caused by increases in inorganic P fractions and less so, for
soils that store most of their P (>84%) in organic forms.
4. Conclusions and implications for management
This study suggests that the physicochemical and P characteristics, P fractions, and P indices of dairy pasture wetland soils were
very different from improved and unimproved pasture wetland
soils. These differences were mostly related to soil characteristics
like soil metal content, TC and inorganic P fractions. Typically, the
greatest nutrient concentrations occurred in surface soils in the top
10 cm layer.
Dairy wetland soils had greatest concentrations of soil total
nutrients along with WEP, most of the P fractions, and metals.
Further, dairy wetland soils also had greatest P sorption capacities, probably related to soil metal content. However, using soil P
indices, we found that dairy wetland soils were P saturated relative to other soils and had no ability to retain additional amounts
of P. The legacy of P in soils, particularly soils collected from dairy
pastures wetlands, combined with effective BMPs to reduce nutri-
Author's personal copy
1490
Ed.J. Dunne et al. / Ecological Engineering 37 (2011) 1481–1491
ent loading to receiving systems could indirectly contribute to
a scenario for P impacted soils to release stored P from soil to
the overlying water column. We hypothesize that this scenario is
temporary, until achieving a new long-term equilibrium between
underlying soil and overlying water. Reddy et al. (2011) suggest
that the possibility of internal loading and the legacy of soil-stored
P could confound restoration activities for an unknown amount of
time, with the magnitude of this being dependent on land use.
One approach to decrease dairy wetland soil P saturation and
to increase the soils ability to retain additional loadings and/or to
mitigate the loss of legacy P already stored in soil is to use soil
amendments (Pant et al., 2002; Leader et al., 2008; Bruland et al.,
2009; Malecki-Brown and White, 2009).
To store P in soils on a long-term basis, soils have to accrete and
accumulate organic matter (Rybczyk et al., 2002; Kadlec, 2009).
Wetlands accumulate organic matter and this process is considered
the main long-term soil sink for P (Craft and Richardson, 1993). We
found that both improved and unimproved pasture wetland soils
stored the majority of their P in organic and residual P fractions.
To control and mitigate for P release and increase P storage, it
will be important to undertake active and dynamic management
of wetland water regimes and hydroperiods (Aldous et al., 2005;
Dunne et al., 2007). Increasing wetland hydroperiod contributes to
decreased organic matter decomposition, increases organic matter accretion rates, directly affects soil water content, soil nutrient
content, and nutrient dynamics between underlying soil and overlying water (Aldous et al., 2005; Leeds et al., 2009) along with the
overall character of the wetland biota (Aldous et al., 2005).
The indices we used are typically used for mineral terrestrial,
agricultural soils that have been loaded with inorganic and organic
fertilizers (Nair and Graetz, 2002; Nair et al., 2004) for many years.
We found that these indices worked well for P impacted dairy pasture wetland soils; however, the use of these indices in improved
and unimproved pasture wetland soils, where the majority of P was
in organic forms, may not be appropriate. We suggest that where
the majority of soil P is organic that an additional organic P factor
like FAP, HAP or ResP be incorporated into the indices.
Acknowledgements
This research was funded by the Florida Department of Agricultural and Consumer Services. We acknowledge the help of
University of Florida’s Wetland Biogeochemistry Laboratory staff,
especially Atanu Mukherjee, and Yu Wang for help with laboratory
sampling and analyses. We thank Soil and Water Science Department students (Charles Bohall, Cory Catts, and Jason Neumann) for
help with fieldwork.
References
Aldous, A., McCormick, P., Ferguson, C., Graham, S., Craft, C., 2005. Hydrologic regime
controls soil phosphorus fluxes in restoration and undisturbed wetlands. Restor.
Ecol. 13, 341–347.
Aldous, A.R., Craft, C.B., Stevens, C.J., Barry, M.J., Bach, L.B., 2007. Soil phosphorus
release from a restoration wetland, Upper Klamath Lake, Oregon. Wetlands 27,
1025–1035.
Andersen, J.M., 1976. An ignition method for determination of total phosphorus in
lake sediments. Water Res. 10, 329–331.
Aye, T.M., Nguyen, M.L., Bolan, N.S., Hedley, M.J., 2006. Phosphorus in soils of riparian
and non-riparian wetland and buffer strips in the Waikato area, New Zealand.
New Zealand J. Agric. Res. 49, 349–358.
Bostic, E.M., White, J.R., 2007. Soil phosphorus and vegetation influence on wetland
phosphorus release after simulated drought. Soil Sci. Soc. Am. J. 71, 238–244.
Braskerud, B.C., 2002. Factors affecting phosphorus retention in small constructed
wetlands treating agricultural non-point source pollution. Ecol. Eng. 19, 41–61.
Bridgham, S.D., Johnston, C.A., Schubauer-Berigan, J.P., Weishampel, P., 2001. Phosphorus sorption dynamics in soils and coupling with surface and pore water in
riverine wetlands. Soil Sci. Soc. Am. J. 65, 577–588.
Bruland, G.L., Richardson, C.J., 2006. An assessment of the phosphorus retention
capacity of wetlands in the Painter Creek Watershed, Minnesota, USA. Water
Air Soil Pollut. 171, 169–184.
Bruland, G.L., Richardson, C.J., Daniels, W.L., 2009. Microbial and geochemical
responses to organic matter amendments in a created wetland. Wetlands 29,
1153–1165.
Cheesman, A.W., Dunne, E.J., Turner, B.L., Reddy, K.R., 2010. Soil phosphorus forms
in hydrologically isolated wetlands and surrounding pasture uplands. J. Environ.
Qual. 39, 1517–1525.
Chrysostome, M., Nair, V.D., Harris, W.G., Rhue, R.D., 2007. Laboratory validation of
soil phosphorus predictions for use in risk assessment. Soil Sci. Soc. Am. J. 71,
1564–1569.
Corstanje, R., Reddy, K.R., 2004. Response of biogeochemical indicators to a drawdown and subsequent flood. J. Environ. Qual. 33, 2357–2366.
Corstanje, R., Portier, K.M., Reddy, K.R., 2009. Discriminant analysis of biogeochemical indicators of nutrient enrichment in Florida wetlands. Eur. J. Soil Sci. 60,
974–981.
Craft, C.B., Richardson, C.J., 1993. Peat accretion and phosphorus accumulation
along a eutrophication gradient in the northern Everglades. Biogeochemistry
22, 133–156.
D’Angelo, E.M., Reddy, K.R., 1994. Diagenesis of organic-matter in a wetland receiving hypereutrophic lake water. 1. Distribution of dissolved nutrients in the soil
and water column. J. Environ. Qual. 23, 928–936.
D’Angelo, E.M., 2005. Phosphorus sorption capacity and exchange by soils from
mitigated and late successional bottomland forest wetlands. Wetlands 25,
297–305.
DeBusk, W.F., Newman, S., Reddy, K.R., 2001. Spatio-temporal patterns of soil phosphorus enrichment in Everglades Water Conservation Area 2A. J. Environ. Qual.
30, 1438–1446.
Dolan, T.J., Bayley, S.E., Zoltek, J., Hermann, A.J., 1981. Phosphorus dynamics of
a Florida fresh-water marsh receiving treated waste-water. J. Appl. Ecol. 18,
205–219.
Dunne, E.J., Reddy, K.R., Clark, M.W., 2006a. Phosphorus release and retention by
soils of natural isolated wetlands. Int. J. Environ. Pollut. 28, 496–516.
Dunne, E.J., Reddy, R., Clark, M.W., 2006b. Biogeochemical indices of phosphorus
retention and release by wetland soils and adjacent stream sediments. Wetlands
26, 1026–1041.
Dunne, E.J., Smith, J., Perkins, D.B., Clark, M.W., Jawitz, J.W., Reddy, K.R., 2007. Phosphorus storages in historically isolated wetland ecosystems and surrounding
pasture uplands. Ecol. Eng. 31, 16–28.
Ekholm, P., Turtola, E., Gronroos, J., Seuri, P., Ylivainio, K., 2005. Phosphorus loss
from different farming systems estimated from soil surface phosphorus balance.
Agric. Ecosyst. Environ. 110, 266–278.
Fisher, M.M., Reddy, K.R., 2001. Phosphorus flux from wetland soils affected by longterm nutrient loading. J. Environ. Qual. 30, 261–271.
Flaig, E.G., Havens, K.E., 1995. Historical trends in the Lake Okeechobee ecosystems
I. Land use and nutrient loadings. Archiv für Hydrobiologie Supplement 107,
1–24.
Flaig, E.G., Reddy, K.R., 1995. Fate of phosphorus in the Lake Okeechobee Basin,
Florida, USA: overview and recommendations. Ecol. Eng. 5, 127–143.
Gathumbi, S.M., Bohlen, P.J., Graetz, D.A., 2005. Nutrient enrichment of wetland and
sediments in subtropical pastures. Soil Sci. Soc. Am. J. 69, 539–548.
Havens, K.E., James, R.T., 2005. The phosphorus mass balance of Lake Okeechobee,
Florida: implications for eutrophication management. Lake Reservoir Manage.
21, 139–148.
Heckrath, G., Bechmann, M., Ekholm, P., Ulén, B., Djodjic, F., Andersen, H.E., 2008.
Review of indexing tools for identifying high risk areas of phosphorus loss in
Nordic catchments. J. Hydrol. 349, 68–87.
Hiscock, J.G., Thourot, C.S., Zhang, J., 2003. Phosphorus budget–land use relationships for the northern Lake Okeechobee watershed, Florida. Ecol. Eng. 21, 63–74.
Ivanoff, D.B., Reddy, K.R., Robinson, S., 1998. Chemical fractionation of organic phosphorus in selected histosols. Soil Sci. 163, 36–45.
Kadlec, R.H., 2009. Wastewater treatment at the Houghton lake wetland: soils and
sediments. Ecol. Eng. 35, 1333–1348.
Kronvang, B., Jeppesen, E., Conley, D.J., Søndergaard, M., Larsen, S.E., Ovesen, N.B.,
Carstensen, J., 2005. Nutrient pressures and ecological responses to nutrient
loading reductions in Danish streams, lakes and coastal waters. J. Hydrol. 304,
274–288.
Leader, J.W, Dunne, E.J., Reddy, K.R., 2008. Phosphorus sorbing materials: sorption
dynamics and physicochemical characteristics. J. Environ. Qual. 37, 174–181.
Leeds, J.A., Garrett, P.B., Newman, J.M., 2009. Assessing impacts of hydropattern
restoration of an overdrained wetland on soil nutrients, vegetation, and fire.
Restor. Ecol. 17, 460–469.
Leibowitz, S.G., 2003. Isolated wetlands and their functions: an ecological perspective. Wetlands 23, 517–531.
Leibowitz, S.G., Nadeau, T.L., 2003. Isolated wetlands: state-of-the-science and
future directions. Wetlands 23, 663–684.
Malecki-Brown, L.M., White, J.R., 2009. Effect of aluminum-containing amendments
on phosphorus sequestration of wastewater treatment wetland soil. Soil Sci. Soc.
Am. J. 73, 852–861.
Maynard, J.J., O’Geen, A.T., Dahlgren, R.A., 2009. Bioavailability and fate of phosphorus in constructed wetlands receiving agricultural runoff in the San Joaquin
Valley, California. J. Environ. Qual. 38, 360–372.
McCarty, G.W., McConnell, L.L., Hapernan, C.J., Sadeghi, A., Graff, C., Hively, W.D.,
Lang, M.W., Fisher, T.R., Jordan, T., Rice, C.P., Codling, E.E., Whitall, D., Lynn, A.,
Author's personal copy
Ed.J. Dunne et al. / Ecological Engineering 37 (2011) 1481–1491
Keppler, J., Fogel, M.L., 2008. Water quality and conservation practice effects in
the Choptank River watershed. J. Soil Water Conserv. 63, 461–474.
McCune, B., Grace, J.B., 2002. With Contributions from D.L. Urban. Analysis of Ecological Communities. MjM Software Design, Glenden Beach, Oregon.
McKeague, J.A., Day, J.H., 1966. Dithionite and oxalate extractable Fe and Al as aids
in differentiating various classes of soils. Can. J. Soil Sci. 46, 3–22.
McKee, K.A., 2005. Predicting soil phosphorus storage in historically isolated wetlands within the Lake Okeechobee priority basins. Unpublished M.Sc. Thesis.
University of Florida, Gainesville, FL.
Merriman, K.R., Gitau, M.W., Chaubey, I., 2009. Tools for estimating best management practice effectiveness in Arkansas. Appl. Eng. Agric. 25, 199–213.
Mukherjee, A., Nair, V.D., Clark, M.W., Reddy, K.R., 2009. Development of indices to
predict phosphorus release from wetland soils. J. Environ. Qual. 38, 878–886.
Nair, V.D., Graetz, D.A., 2002. Phosphorus saturation in spodosols impacted by
manure. J. Environ. Qual. 31, 1279–1285.
Nair, V.D., Harris, W.G., 2004. A capacity actor as an alternative to soil test phosphorus in phosphorus risk assessment. New Zealand J. Agric. Res. 47, 491–497.
Nair, V.D., Portier, K.M., Graetz, D.A., Walker, M.L., 2004. An environmental threshold
for degree of phosphorus saturation in sandy soils. J. Environ. Qual. 33, 107–113.
Noe, G.B., Hupp, C.R., 2009. Retention of riverine sediment and nutrient loads by
coastal plain floodplains. Ecosystems 12, 728–746.
Pant, H.K., Reddy, K.R., 2003. Potential internal loading of phosphorus in a wetland
constructed in agricultural land. Water Res. 37, 965–972.
Pant, H.K., Reddy, K.R., Spechler, R.M., 2002. Phosphorus retention in soils from a
prospective constructed wetland site: environmental implications. Soil Sci. 167,
607–615.
Reddy, K.R., O’Connor, G.A., Gale, P.M., 1998. Phosphorus sorption capacities of wetland soils and stream sediments impacted by dairy effluent. J. Environ. Qual. 27,
438–447.
Reddy, K.R., Kadlec, R.H., Flaig, E., Gale, P.M., 1999. Phosphorus retention in streams
and wetlands: a review. Crit. Rev. Env. Sci. Technol. 29, 83–146.
1491
Reddy, K.R., Wetzel, R.G., Kadlec, R.H., 2005. Biogeochemistry of Phosphorus in Wetlands. Phosphorus: Agriculture and the Environment, Agronomy Monograph No.
46. American Society of Agronomy, Madison, WI, pp. 263–316.
Reddy, K.R., Newman, S., Osborne, T.Z., White, J.R., Fitz, H.C., 2011. Phosphorus
cycling in the Everglades ecosystem: legacy phosphorus implications for management and restoration. Crit. Rev. Env. Sci. Technol. 41, 149–186.
Rybczyk, J.M., Day, J.W., Conner, W.H., 2002. The impact of wastewater effluent
on accretion and decomposition in a subsiding forested wetland. Wetlands 22,
18–32.
Savvides, A., Corstanje, R., Baxter, S.J., Rawlins, B.G., Lark, R.M., 2010. The relationship
between diffuse spectral reflectance of the soil and its cation exchange capacity
is scale-dependent. Geoderma 154, 353–358.
South Florida Water Management District (SFWMD), 2003. Land Use GIS Layer.
South Florida Water Management District, West Palm Beach, FL.
Sharpley, A., 1999. Global issues of phosphorus in terrestrial ecosystems. In: Reddy,
K.R., O’Connor, G.A., Schelske, C.L. (Eds.), Phosphorus Biogeochemistry in Subtropical Ecosystems. Lewis Publishers, Boca Raton, FL, pp. 15–46.
Søndergaard, M., Jensen, J.P., Jeppesen, E., 2005. Seasonal response of nutrients to
reduced phosphorus loading in 12 Danish lakes. Freshwater Biol. 50, 1605–1615.
Tiner, R.W., 2003. Geographically isolated wetlands of the United States. Wetlands
23, 494–516.
Turner, B.L., Cade-Menun, B.J., Condron, L.M., Newman, S., 2005. Extraction of soil
organic phosphorus. Talanta 66, 294–306.
United States Environmental Protection Agency (USEPA), 1993. Methods for
Chemical Analysis of Water and Wastes. Environmental Monitoring, Support
Laboratory, Cincinnati, OH.
Whigham, D.F., Jordan, T.E., 2003. Isolated wetlands and water quality. Wetlands
23, 541–549.
Winter, T.C., LaBaugh, J.W., 2003. Hydrologic considerations in defining isolated
wetlands. Wetlands 23, 532–540.