Citation
Enhanced Retention of Phosphorus Applied as Flushed Dairy Manure

Material Information

Title:
Enhanced Retention of Phosphorus Applied as Flushed Dairy Manure
Creator:
Malek, Aaron J
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (54 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Interdisciplinary Ecology
Committee Chair:
Harris, Willie G.
Committee Co-Chair:
Wilkie, Ann C.
Committee Members:
Nair, Vimala D.
Rhue, Roy D.
Graduation Date:
12/14/2007

Subjects

Subjects / Keywords:
Adsorption ( jstor )
Dairy manure ( jstor )
Leaching ( jstor )
Manure ( jstor )
Phosphorus ( jstor )
Porosity ( jstor )
Reagents ( jstor )
Soil horizons ( jstor )
Soils ( jstor )
Sorption ( jstor )
Interdisciplinary Ecology -- Dissertations, Academic -- UF
leaching, phosphorus, soils
City of Tavares ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Interdisciplinary Ecology thesis, M.S.

Notes

Abstract:
Leaching of phosphorus (P) is a major concern in many areas of the world, especially in environments with karst topography that contain sensitive and valuable aquifers and springs. In one such area, North Florida of the SE United States, dairy farming is prevalent, and in such operations it is common to spray crop fields with flushed dairy manure (FDM) as a means of nutrient recycling. We investigated the fate of P in three sandy soils that typify and encompass large areas of the region. At the University of Florida Dairy Research Center, anaerobic digestion of FDM is currently practiced and provides many benefits including reduction of organic matter (OM) and mineralization of nutrients, including P. Three different P amendments were applied to soil columns in a randomized complete block design: raw FDM, digested FDM, and inorganic P. Since OM is known to form complexes and compete for sorption sites with P, thereby reducing P sorption, it was hypothesized that retardation of P would be least in soils receiving P as raw FDM and most in soils receiving inorganic P. Soil columns were kept under 20 cm of suction to promote unsaturated flow and amendments were applied at one half of the soil pore volume. Leachate was collected per pore volume and analyzed colorimetrically for total P (TP) and dissolved reactive P (DRP). We found little evidence of OM transport of P in columns receiving both raw and digested FDM, and through 30 pore volumes, very little leaching of P at all. The soils receiving inorganic P retarded leaching the least, and the extent of retardation was related to the soils inherent P sorbing capacity through Al and Fe oxide content. ANOVA output revealed significant differences in P accumulation in the soils among P amendments and soils, as well as a P amendment * soil interaction. Duncan?s Multiple Range Test showed the P accumulated in soils receiving inorganic P to be significantly less than those receiving raw or digested FDM. Precipitation with Ca and Mg contained in the FDM, entrapment of particulate P, and immobility of OM due to soil chemistry are the suspected explanations for the lack of P breakthrough in FDM amended columns. The precipitated P is not expected to be stable and will leach into the deeper soil horizons. This study highlights the differences in P behavior in soils when delivered through different mediums and the complexity of the soil chemistry regarding P when FDM is amended to soils. It also demands further research in this area to better understand the processes and consequences of P leaching where similar practices are employed. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (M.S.)--University of Florida, 2007.
Local:
Adviser: Harris, Willie G.
Local:
Co-adviser: Wilkie, Ann C.
Statement of Responsibility:
by Aaron J Malek.

Record Information

Source Institution:
UFRGP
Rights Management:
Copyright Malek, Aaron J. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
663112687 ( OCLC )
Classification:
LD1780 2007 ( lcc )

Downloads

This item has the following downloads:


Full Text





ENHANCED RETENTION OF PHOSPHORUS APPLIED AS FLUSHED DAIRY MANURE


By

AARON MALEK
















A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF
FLORIDA INT PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2007









CHAPTER 1
INTTRODUCTION

Background

Phosphorus (P) is an economically important input in both crop and livestock production

systems (Hansen et al. 2002). It also poses an environmental risk if it migrates elsewhere in the

environment. Increased concentrations of P in inland and coastal waters can lead to problems

associated with eutrophication. Therefore, great importance has been given to determine the

behavior and movement of P in the environment. Phosphorus is transported from agricultural

fields via surface runoff and leaching. Leaching of P is especially important in sandy soils and

even more important in soils overlying sensitive and valuable aquifers. As part of nutrient

management strategies, it is common practice on many dairy farms to spray fields with flushed

manure in order to improve crop production and pasture land. However, many times high P

levels occur in soils as a result of N:P imbalance in the wastewater.

Studies by lyamuremye et al. (1996) and others highlight the importance of the

interaction of P with organic matter. Dissolved organic matter may compete with P for sorption

sites in soil, displace sorbed P, and mediate movement by sorption with P itself. Particulate

organic matter may physically block sorption sites, also promoting leaching of P. It follows that

the application of inorganic phosphate may lead to less leaching of P than when applied in the

presence of organic. Flushed dairy manure (FDM) wastewater has substantial amounts of

suspended particulate and dissolved organic matter. However, innovative techniques in treating

FDM are increasingly used to improve farm operations in various ways. Fixed-filmed anaerobic

digestion of FDM is one such technique, and it significantly lowers and alters the organic matter

content which may decrease leaching of P (Wilkie, 2004). Some of the benefits of anaerobic

digesters include decreasing potential adverse effects the FDM may have on animal and plant life










through macropores 1mm or greater with negligible P sorption to pore walls. When macropores

were absent, no measurable P was transported through the soil column. The macropores were

disrupted from plowing which promoted matrix flow and reduced P leaching.

The literature, to date, explains many of the factors concerning the potential for P

mobility in soils. With regards to sandy soils, studies have shown variable ability to uptake P.

Some of the sandy soils had little capacity while others almost completely prevented leaching.

This is due, in part, to the inherent P sorbing compounds present in the soil, such as Fe and Al

oxides. When organic amendments are added to the soil, complexes that mediate transfer of P

can form. The extent that this occurs varies among soils and also depends on the P source.

Organic sources of P, such as raw and digested FDM, can have different effects on P mobility

than other organic sources. The consequences of excessive P leaching, such as eutrophication,

require an understanding of all the contributing factors. Further research of P movement in

various soils and application methods will aid in the development of management practices that

both aid in agricultural production and prevent water pollution.












sysem Hav arows iniae lwo fuhatradmauesois ) ikn
paror B)Fesalbrs )Flswtrhligtns D atwtrcleto
channel E)- Sadta.F ehncl eaao.G eietainbsn )smln
Pit. )Pimr traepnd Wlie 04













so, .A so

40 -I w 40








O 5 10 15 20 25 301 35 0 5 10 15 20 25 30) 35
Pore Volume Pore Volume

tPhosphate
so, G-+-Digested FDM
-a-RawvFDM









0 5 10 15s 20 25 30 35
Pore Volume


Figure 3-4. Dissolved reactive P (DRP) concentrations in leachate for each pore volume. With
regards to flushed dairy manure (FDM) amended columns, concentrations DRP and
total P are similar which suggests that nearly all total P leaching through the columns
is DRP. A) Millhopper DRP. B) Tavares DRP. C) Orsino DRP.













70 -
60 -
a 50 -
E


a 20 -
10 -


y = 0i762x -13.55
RS= 0.896


30 40 50 60 70 80 90 100
RPA

Figure 3-1. Correlation between total oxalate extracted Fe and Al in molar concentrations and
RPA values of the soils. Data points include soils Millhopper E2, Orsino, Millhopper
El, Tavares E2, and Tavares El in order of ascending RPA (Table 3-4).


Table 3-4. Characterization of soils collected. (n


TP RPA Fe ?
mg kg (%) mg kg '


Al Jf Total Fe &
mg kg Al
mmol kg'
857 41.5
1380 65.7
1008 52.7
919 39.0
511 24.7


Sand g Silt Clay
kg g kg g kg'


Orsino Bw
Tavares El
Tavares E2
Millhopper El
Millhopper E2


544
811
856
279
324


980
950
950
940
950


Jf oxalate extracted
Jft clay was lost, so only sand is reported


Table 3-5. Sand size distribution among soils collected and pore volume of soil horizons used in
the experiment.


% Ver, % Coarse % Medium
Coarse
.2 3.0 22.6
.3 7.0 44.5
.3 6.1 42.4
.3 5.8 43.9
.3 5.7 45.9


% Fine % Very Fine Pore volume at 20
cm suction (mL)
66.7 5.3 5.0
36.2 6.9
37.7 8.5 4.2
40.3 3.5
39.3 4.1 4.7


Orsino
Tavares El
Tavares E2
Millhopper El
Millhopper E2









may be mobile depending on pH, ionic strength, and the ratio of monovalent to divalent cations.

They state low ionic strengths and high ratios of monovalent to divalent cations promote organic

matter mobility. Also, when they added Al and Fe to soils, the amount of these complexes

(defined as high molecular mass molybdate reactive P), increased, but when they just added

inorganic P, no increase was observed. This indicates that the metal must bond to the OM prior

to the P. The soils in their experiment had low ionic strengths, low Ca concentrations, and high

Na/Ca ratios, whereas the soils in this experiment certainly have the opposite conditions. This

may explain why there was little difference in the TP and DRP concentrations in the leachates of

this experiment. Hesketh et al. (2001) also investigated colloid transport of P using lysimeters

with soils amended with pig slurry and found very little evidence of it; 1% of P applied was

leached. The combination of precipitation and the maintenance of high divalent cation

concentration in the soil may contribute to the prevention of immediate leaching of P. If FDM

amendments are suspended, the pH and amount of divalent cations would decrease with rainfall,

perhaps increasing the mobility of P sorbed to OM and precipitated as Ca-P.

An interesting factor in this scenario is the bond strength of the OM-M versus the M-P

bond. If OM-M-P complexes leach into E horizons with significant amounts of Al/Fe oxide sand

coatings, they could sorb as a solid phase. Hens and Merkx (2001) state that metastable OM-M-P

complexes will destabilize with decreasing pH, a condition that would be found in the lower

horizons of acidic soils. If the OM-M bond hydrolyzes more easily than the M-P bond, in some

instances, P may be occluded without decreasing the sorption capacity (assuming high

concentrations of available P bonding metals). A series of reactions may occur where P and

metals may build upon each other when OM-M bonds destabilize:

M-sand + P & P-M-sand,










Karathanasis, A.D. and D.M.C. Johnson. 2006. Subsurface transport of Cd, Cr, and Mo mediated
by biosolid colloids. Sci. Total Environ. 354:157-169.

Lehmann, J., Z. Lan, C. Hyland, S. Sato, D. Solomon, and Q.M. Ketterings. 2005. Long-term
dynamics of phosphorus forms and retention in manure-amended soils. Environ. Sci.
Technol. 39:6672-6680.

Leytem, A.B., R.L. Mikkelsen, and J.W. Gilliam. 2002. Sorption of organic phosphorus
compounds in Atlantic coastal plain soils. Soil Sci. 167:652-658.

Lilienfein, J., R.G. Qualls, S.M. Uselman, and S.D. Bridgham. 2004. Adsorption of dissolved
organic and inorganic phosphorus in soils of a weathering chronosequence. Soil Sci. Soc.
Am. J. 68:620-628.

McKeague, J.A., and D.H. Day. 1966. Dithionite and oxalate extractable Fe and Al as aids in
differentiating various classes of soils. Canadian J. Soil. Sci. 46: 13-22.

Makris, K.C., W.G. Harris, G.A. O'Connor, and H. El-Shall. 2005. Long-term phosphorus
effects on evolving physicochemical properties of iron and aluminum hydroxides. J.
Colloid Interf. Sci. 287:552-560.

Motoshita, M., T. Komatsu, P. Moldrup, L.W. de Jonge, N. Ozaki, and T. Fukushima. 2003. Soil
constituent facilitated transport of phosphorus from a high-P surface soil. Soils and
Foundations 43:105-114.

Murphy, J., and J.P. Riley. 1962. A modified single solution method for determination of
phosphate in natural waters. Anal. Chim. Acta 27:31-36.

Nair, V.D., R.R. Villapando, and D.A. Graetz. 1999. Phosphorus retention capacity of the spodic
horizon under varying environmental conditions. J. Environ. Qual. 28:1308-1313.

Novak, J.M. and D.W. Watts. 2004. Increasing the phosphorus sorption capacity of southeastern
Coastal Plain soils using water treatment residuals. Soil Sci. 169:206-214.

Pardo, P., J.F. Lopez-Sanchez, and G.Rauret. 2003. Relationships between phosphorus
fractionation and maj or components in sediments using the SMT harmonized extraction
procedure. Anal. Bioanal. Chem.376:248-254.

Phillips, I.R. 2002. Phosphorus sorption and nitrogen transformation in two soils treated with
piggery wastewater. Aust. J. Soil Res. 40:335-349.

Phillips, I.R. 2002. Nutrient losses from undisturbed soil cores following applications of piggery
wastewater. Aust. J. Soil Res. 40:515-532.

Qualls, R.G., and B.L. Haines. 1991. Geochemistry of dissolved organic nutrients in water
percolating through forest ecosystems. Soil Sci. Soc. Am. J. 55:1112-1123.














Table 3-3. Orsino typical pedon description, taken from the Levy County Soil Survey Report
(Slabaugh et al., 1996). The Orsino Bw horizon used in this study was collected from
a similar Orsino soil in Levy County.
Horizon Depth (cm) Description
A 0-10 Gray (10YR 5/1) Eine sand; weak Eine granular structure; very
friable; many Eine roots, strongly acid; clear wavy boundary.
El 10-20 Very pale brown (10YR 7/2) Eine sand; single grained; loose;
strongly acid; clear wavy boundary.
E2 20-32 White (10YR 8/1) Eine sand; single grained; loose; strongly
acid; abrupt irregular boundary.
Bw and 32-122 Brownish yellow (10YR 6/6) Eine sand (Bw); single grained;
Bh loose; common fine roots; discontinuous lenses of weakly-
cemented dark yellowish brown (10YR 4/4) Eine sand (Bh) that
are 1 to 5 cm thick at the upper contact of the horizon; strongly
acid; gradual wavy boundary.
Bwl 122-147 Light yellowish brown (10YR 6/4) fine sand; few Eine faint
brownish yellow (10YR 6/6) mottles; single grained; loose; few
fine roots; strongly acid, gradual wavy boundary.
Bw2 147-175 Brownish yellow (10YR 6/8) fine sand; single grained; common
fine distinct strong brown (7.5YR 5/8) mottles; loose; few fine
roots; strongly acid; gradual wavy boundary.
C 175-200 White (10YR 8/1) fine sand; few medium distinct yellow
(10YR 7/8 mottles; single grained; loose; moderately acid.









remained suspended. At the time of each addition, the jugs were slightly shaken, in a similar

way, to mix the suspended particles and then a small portion was transferred to a separate beaker.

Each jug was not used once it had become halfway empty. This ensured an approximately equal

consistency of the FDM added to the soil columns that was evident by observation.

Leaching Procedure

Each soil received three treatments: raw FDM, digested FDM, and a solution of KH2PO4.

Based on TP analyses the solution of KH2PO4 WAS prepared at 48 mg L1 P with a 50 mmol KCl

background electrolyte concentration. This ensured roughly equal TP values for all amendments.

Treatments were done in triplicate per soil in a randomized complete block design. To allow for

adequate sorption of P, the rate of addition was 0.5 pore volume of the amendment per addition.

The leachate was collected per whole pore volume.

The rate of the additions was limited by clogging of some columns receiving FDM

amendments. In some cases more than 48 hours elapsed for complete infiltration. The clogging

occurred in the top portion of the columns as the larger particulates settled there and slowed

infiltration. The columns were kept in dark as much as possible to discourage algal growth.

Leachate Analyses

The soil leachates were analyzed for TP and DRP. Total P determination was performed

by the ashing procedure without HCI digestion, as described in the "Total P method

investigation" section. Dissolved reactive P analysis included filtration, dilution if needed, and

the addition of Reagent Bw. Absorbance was taken at 880nm after 2 hours.

Statistical Analyses

An ANOVA was run evaluating the statistical significance of the randomized complete

block design experiment, and Duncan's Multiple Range Test was used to determine significant

differences in the P accumulated in the columns among the three treatments. Both of the









CHAPTER 4
DISCUSSION

Organic matter has been known to promote the leaching of P by competing with or

blocking of sorption sites, displacing P that is sorbed, and mediating transport of P (Agbenin and

Igbokwe, 2006; Andrade et al., 2003; von Wandruszka, 2006; Hens and Merckx, 2001).

However, as evident in the lack of P breakthrough in this study among the FDM treated columns

(Figures 3-3 and 3-4), other processes must play a maj or role in the fate and transport of P as

applied as FDM on sandy soils in the southeastern United States. Sorption to Fe and Al oxide

coatings is a dominant process by which P is fixed in soils similar to these in Florida. Indeed,

there was a strong correlation of the RPA values of the soils used in this experiment and the

amount of oxalate extracted Fe and Al they contained (Figure 3-1). However, other mechanisms

affecting retention of P applied as FDM likely involve components delivered to the soil by the

FDM itself, since retention of P applied as FDM did not vary with the soil relative P adsorption

capacity as it did for the inorganic P treatment. It is unlikely that the lower DRP concentrations

in the FDM could have caused the delayed breakthroughs shown in Figures 3-3 and 3-4, as the

concentrations are not low enough based on an RPA-retardation relation determined for sandy

soils by Rhue et al. (2006). Prospective mechanisms explaining the retention of P added as FDM

are (i) precipitation of P with Ca and Mg and (ii) entrapment of particulate P in the soil column.

The role of Ca in P fixation in soils is well documented in semi-arid and arid regions

where calcareous soils are prevalent (Delgado et al., 2002). This is a maj or problem concerning

farmers, whose fertilizer P becomes unavailable for crops. Many authors have written on the

subj ect and have highlighted the role of organic amendments to increase P availability in the soil

(Agbenin and Igbokwe, 2006; Bennani et al., 2005; Fernandez-Perez et al., 2005; Delgado et al.,

2002; Cong and Merkx, 2005). This would seem to discourage the use of organic amendments,









in associated water bodies by lowering biological oxygen demand (BOD), decreasing fecal

coliforms, and production of methane. It is possible the digested effluent could also be used to

minimize P leaching. This study seeks to determine if there is less retardation of P when it is

applied with organic compared to inorganic application, and if so, whether there is a difference

in retardation between anaerobically digested FDM and raw FDM. The results of this study will

highlight differences, if any, between the movement of inorganic and organic P, and contribute

to the understanding of the role of organic matter in P mobility.

Hypotheses

The main hypothesis at the outset of this research was that organic matter in the FDM

wastewater would compete with P for sorption sites in the soil, displace P on sorption sites, and

mediate movement by sorption with P. Thus, soils applied with inorganic phosphate would retard

P mobility the greatest. Because the digested FDM wastewater has less organic matter and

organic P than the raw FDM wastewater, it should have greater retardation of P movement. The

actual results of this study, however, documented behavior that directly contrasted with these

initial expectations.

Obj ectives

Determine P leaching potential inorganic and total P when soil is amended with raw
FDM wastewater, anaerobically digested FDM wastewater, and inorganic phosphate.

Determine the relative P sorption capacity of sandy soil materials used in the experiment.

Literature Review

The movement of P through soils has been examined in many studies (Sharpley et al.

2003). The interactions and transformations of P within the soil profile are important in

understanding the leaching risk of P. Studies examining different forms of P and P sources have

also highlighted important differences in reactivity and leaching potential. Research has shown










in leachate from that which was added to the columns. The P delivered as inorganic P

accumulated differently and relates to the RPA (Figure 3-7), which was found to be well

correlated with the amount of oxalate extractable Fe and Al oxides (Figure 3-1). Table 3-7

shows the accumulation of P in the soil columns only differ significantly between the inorganic P

amendments and the FDM amendments. Differences between soils were not tested statistically

due to the soil x amendment interaction (Tables 3-7 and 3-8).

Table 3-1. Tavares soil description


Horizon Depth (cm)
A 0-15


Description
Dark grayish brown (10YR 4/2) sand, clear boundary.
Brown (10YR 5/3) sand, gradual boundary.
Brown (10YR 5/3) sand, gradual boundary.
Light yellowish brown (10YR 6/4) sand, gradual boundary.
Pale brown (10YR 6/3) sand, redox concentrations and
depletions, gradual boundary.
Very pale brown (10YR 7/3) sand, redox concentrations and
depletions, gradual boundary.

soil description
Description
Very dark grayish brown (10YR 3/2) sand, clear boundary.
Brown (10YR 4/3) sand, clear boundary.
Yellowish brown (10YR 5/4) sand, gradual boundary.
Brown (10YR 5/3) sand, gradual boundary.
Pale brown (10YR 6/3) sand, gradual boundary.
Light gray (10YR 7/2) sand, redox concentrations and
depletions, red matrix, clear boundary.
Brown (7.5YR 5/4) sandy loam, redox concentrations and
depletions, clear boundary.
Light gray (10YR 7/1) sandy clay loam, redox concentrations
and depletions, gradual boundary.
Light gray (10YR 7/1) sandy clay, redox concentrations
and depletions.


AE
El
E2
E3

E4


Table 3-2.
Horizon
A
AE
El
E2
E3
E4

Bt

Btgl

Btg2


15-50
50-75
75-95
95-120

120-200


Millhopper
Depth (cm)
0-14
14-29
29-52
52-83
83-105
105-118

118-140

140-175

175-189









leaching in structured clay soils. Leaching of total P and dissolved reactive P were measured

over three years in undisturbed soil columns. There was no general correlation between P

concentrations and soil test P or P sorption indices of the topsoil. In one soil, where preferential

flow was the dominant water transport pathway, water and P bypassed the high sorption capacity

of the soil, resulting in high P losses. A comparison of two soil textural extremes, a clay loam

and loamy sand, was conducted by van Es et al. (2004). The study took place on farm land with

artificial drainage and liquid manure application. High P leaching losses were measured in the

clay loam as soon as drain lines initiated flow after manure application. Flow weighted mean P

leaching losses on clay loam plots averaged 39 times higher than those on loamy sand plots.

Preferential flow was determined to be the main transport mechanism in the clay loam, but the

authors stated P leaching from manure applications on loamy sand soils do not pose

environmental concerns as long as soil P levels remain below saturation levels. This is further

supported by Akhtar et al. (2003) where five soils of differing textures were leached with

synthetic acid rainwater enriched with inorganic and organic P. At low flow rates, P appeared in

the drainage water soon after application of either inorganic or organic P for the silt loam soil.

Soils in which matrix type flow dominated had little or no increase in drainage water P. Elevated

P concentrations in the drainage water could not be explained by the P adsorption strength of the

soils with the possible exception of the sandy loam soil, where the outflow P concentration was

consistently low.

Agricultural practices, such as plowing, affect the porosity of the soil. Geohring et al.

(2001) found that plowing in liquid manure applications before irrigation greatly reduced P

leaching. Column studies using pack soil and artificial macropores were designed to examine the

role of macropore size on P sorption to pore walls. They found that soluble P may be transported



















er I Jy= 0.OB2x- 2.271
2 -1 R'=0.926

1-1



0 20 40 60 BO 100
RPA (%)b
Figure 3-7. Correlation between P accumulated in soil columns, delivered as inorganic P at 48
mg L1 P solution, and soil RPA values. In ascending RPA values, data points include
the soils Millhopper E2, Orsino Bw, and Tavares E2. (p< 0.0001)


































To my Mom




































O 2007 Aaron Malek









both raw and digested FDM, and through 30 pore volumes, very little leaching of P at all. The

soils receiving inorganic P retarded leaching the least, and the extent of retardation was related to

the soils inherent P sorbing capacity through Al and Fe oxide content. ANOVA output revealed

significant differences in P accumulation in the soils among P amendments and soils, as well as a

P amendment soil interaction. Duncan's Multiple Range Test showed the P accumulated in

soils receiving inorganic P to be significantly less than those receiving raw or digested FDM.

Precipitation with Ca and Mg contained in the FDM, entrapment of particulate P, and immobility

of OM due to soil chemistry are the suspected explanations for the lack of P breakthrough in

FDM amended columns. The precipitated P is not expected to be stable and will leach into the

deeper soil horizons. This study highlights the differences in P behavior in soils when delivered

through different mediums and the complexity of the soil chemistry regarding P when FDM is

amended to soils. It also demands further research in this area to better understand the processes

and consequences of P leaching where similar practices are employed.












TABLE OF CONTENTS


page

ACKNOWLEDGMENTS .............. ...............4.....


LIST OF TABLES .........__.. ..... .__. ...............7....


LIST OF FIGURES .............. ...............8.....


AB S TRAC T ......_ ................. ............_........9


CHAPTER


1 INTRODUCTION ................. ...............11.......... ......


Background ................. ...............11.................
Hypotheses............... ...............1
Obj ectives ................. ...............12.......... .....
Literature Review .............. ...............12....


2 MATERIALS AND METHODS .............. ...............20....


Soil Selection and Sampling ................. ...............20................
Soil A nalyses ................ .. .......... .. .................2
Oxalate-Extractable Iron (Fe) and Aluminum (Al) ................................... 2
Total P (TP) of Soils.............. .............. ...............2
Relative Phosphorus Retention Capacity (RPA) ...._.. ................ ........_.... ....22
Particle Size Distribution Analysis (PSDA) of Soils .............. ...............23....
Flushed Dairy Manure Analyses ................ ... .................. .......2
Total P Method Investigation HCI Digestion vs Only Ashing .............. ...................25
Dissolved Reactive Phosphorus (DRP) ....._.. ............... ............_ ....... 2
Chemical Oxygen Demand (COD) .............. ...............26....
Soil Column Experimental Procedures. .........._.._....._.. ...............26...
Column Construction................ .............2
Pore Volume Determination............... .............2

Background P ......................... ... .... .. .........2
Flusded Dairy Manure (FDM) Storage and Sampling .............. ...............27....
Leaching Procedure .........._.... ......... ...............28.....
Leachate Analyses .........._.... ......... ...............28.....
Statistical Analyses............... ...............28

3 RE SULT S .............. ...............3 0....


Soil Descriptions............... ..............3
Soil Characterization .............. ........ .... ..........
Flushed Dariy Manure (FDM) Characterization .............. ...............31....
Analyses of P in Leachate. .........._._ ...._.... ...............31 ....










preceding tests utilized SAS software. A regression analysis was used to demonstrate the

significance of the relationship of P accumulated in the columns receiving P as PO4 and the RPA

value of the soil.


figure z-1. columns containmng Zug or sonl were itept unaer zu cm or suction utinzing
Erlenmeyer flasks to ensure unsaturated flow of leachate, which was collected in
glass vials.









determining adsorption of P. Novak and Watts (2004) increased the sorption capacity of a sandy

soil several fold by incorporating water treatment residuals that were flocculated with liquid

alum (Al2(SO4)3). Freese et al. (1992) found phosphate adsorption in German soils to be

predominantly related to the amount of amorphous Fe and Al oxides. Also, Phillips (2002) found

leaching to be related to the availability of Fe oxides on soil colloids. Al-bound P was the main

form of P when Elrashidi et al. (2001) studied accumulation of P in Florida sandy soils (Myakka,

Zolfo, and Adamsville). The P moved through the soil profile, but the low solubility of Al-

phosphates (Al-P) in the acidic saturated zone of these soils prevented infiltration into

groundwater. Ca-bound P (Ca-P) is also pH sensitive but in some forms they are extremely

insoluble and considered largely as unavailable (Brady and Weil, 2002; Pardo et al., 2003). Zhou

et al. (1997) found the high adsorption capacity of the Spodosol Bh horizon a result of

aluminum-organic matter complexes. Theses complexes can form in the soil profile and affect

the P leaching potential as they release P more readily than inorganic metal oxides.

Many studies have examined the interactions of P and other soil constituents, especially

comparing these interactions when the P form is inorganic or organic. Most have found

adsorption of inorganic P greater than organic P, but not all. In fact, Leytem et al. (2002) found

organic P forms to be preferentially adsorbed over inorganic, ortho-P. They studied the behavior

of four organic P compounds: three nucleotides, ATP, ADP, AMP, and IHP inositoll

hexaphosphate) while KH2PO4 WAS used as an inorganic reference. All the organic P compounds

had greater adsorption than KH2PO4 On Blanton Sand and Cecil sandy clay loam. In the

Belhaven sandy loam, IHP had the greatest adsorption followed by KH2PO4 and the nucleotides.

When Lilienfein et al. (2004) studied preferential adsorption of organic and inorganic P, the

organic P already present in the soil was examined, not specific organic P compounds. They









Erlenmeyer flask controls the amount of suction. For this experiment the flasks were held at an

elevation that results in 20 cm of suction (Figure 2-1).

Pore Volume Determination

The mass of soil used in each column for this experiment was 20 g. The pore volume for

each soil was determined by placing 20 g of soil in weighed columns, wetting with 50 mM KC1,

and applying 20 cm of suction. When excess KCl finished passing through the column and any

addition of KCl resulted in immediate elution, the column was weighed again to determine the

pore volume.

Background P

The background P of the soils was determined by elution with 50 mM KC1. The first

background value was collected from leachate from the initial wetting of the soil. The

subsequent background values were taken from leachate obtained by the addition of 0.5 pore

volumes of KCl to the columns at least one day apart. A total of 6 KCl pore volumes were

collected to get steady background P values. Then, 2.5 mL aliquots of the soil leachates were

filtered through 0.45 Cpm filter, placed in test tubes, and dried in an oven overnight at 70 C. A set

of standards: 0. 1 mL of 0, 1, 2, 3, 4, and 5 mg L1 P was made. Each sample and standard

received 5 mL of Reagent Bw, and color developed for two hours at which time absorbance

readings were taken at 880 nm.

Flushed Dairy Manure (FDM) Storage and Sampling

Approximately 1 1.5 liters of both raw FDM and digested FDM were collected for the

experiment. They were collected and stored in 3.8 L jugs at 40C. A procedure was devised to

minimize anticipated clogging of soil columns and to provide a regular consistency of the FDM

amendments when applied to soil columns and when sampled for analytical purposes. The

largest suspended particles were allowed to settle at the bottom of the jugs. The smaller particles


































































__ ~~11_1


L-rLI~


25 30 2


25 30 35 0 5 10 15 20
Pore Volume






16-
14-









25 30 2 0 5 10 5 20
Pore Volume


mA
zs
Is
14

j
18
E 8
6
4
Irl
2
B


2~~L~
18
15
14
12
j
18
Es


F;i 21 1~IC~

15 15 25 25 3II 55 B 5 IB 15 25 25 3[1 35
Pore Volume Pore Volume


18
16
14


"`i


25 30 2


5 110 15 20
Porer Volume


~Pt~tr


14 -


t~
~L _.L~ft~


O5 101 15 20
Pore Volume


Figure 3-5. P in leachates for flushed dairy manure (FDM) treatments, shown as both total P

(TP) and dissolved reactive P (DRP). With the exception of the beginning pore
volumes, this illustrates that P in the leachate is largely comprised of DRP. A)

Millhopper raw FDM: TP vs. DRP. B) Millhopper digested FDM: TP vs. DRP. C)

Orsino raw FDM: TP vs. DRP. D) Orsino digested FDM: TP vs. DRP. E) Tavares

raw FDM: TP vs. DRP. F) Tavares digested FDM: TP vs. DRP.









FDM than for the raw FDM, considering that DRP comprised a higher proportion of TP for the

former. The benefit of digested FDM over raw FDM in spray field applications may be more

linked with factors other than leaching of P, such as more plant available P and odor and

pathogen reduction. The higher proportion of plant available P in the digested FDM would be

preferable because more of the P would be available for immediate uptake through the crops

when the FDM is applied.

These soils have a predictable sorption capacity when P is delivered as KH2PO4 and is

largely a function of the RPA (Fig 3-7). The soils can quickly reach their sorption capacity and

then offer little resistance to P leaching and leachate P concentrations could be very high.

However, the Al/Fe-P is more stable than the Ca-P minerals formed under the influence of

manure (Hansen and Strawn, 2003; Makris et al. 2005). Depending on the environmental,

agricultural, and practical conditions, one method of P fertilization may be favored over another.

Concerning leaching of P in these soils, when at similar P concentrations, the largest apparent

difference between P delivered as FDM and P delivered as KH2PO4 is that when irrigating with

FDM, the surface horizon acts as an expanding, but leaky bucket with regards to containing P;

that is, the P is accumulated in the surface horizon but steadily leaches at a slow rate. However,

the soil acts as a "fixed" bucket when using an inorganic P fertilizer. P progressively travels

down the soil profile becoming more permanently fixed than metastable Ca-P, but reaches its

inherent P capacity as Al/Fe-P, and then the soil offers no more resistance to P leaching. As

stated above, more analyses of these experimental soils could provide useful information

concerning P fate and environmental risk.

Future Studies

Further study of the soil columns could evaluate the evidence of Ca-P precipitation, the

extent to which it occurs, and its relevant properties. An analysis, such as x-ray diffraction, could









dark, cool place. Reagent Bs was made fresh from a stock solution of Reagent As for each

analysis. To make Reagent Bs, 1.056 g of ascorbic acid was dissolved in 200 mL of Reagent As.

Reagent Bs was added in 4 mL aliquots to test tubes containing soil extractant solutions

or standards. Then, 10 mL of deionized H20 was added while turning the tube to rinse the sides.

Blue color developed in the test tubes commensurate with the amount of P they contained. The

absorbance readings were taken on a spectrophotometer at 880 nm wavelength after color

developed for thirty minutes. The standard curve was accepted if the R2 was at least 0.997. The

absorbance readings of the soil samples were interpolated on the standard curve to determine P

mg L^1

Relative Phosphorus Retention Capacity (RPA)

The RPA is a procedure developed to provide a quick practical measure of the P sorption

capacity of sandy soils (Rhue et al, 1994). This analysis used Reagent A and Reagent B that

differs somewhat from the Reagent A and Reagent B that was used in determining TP of the

soils. For convenience, the reagents used in the TP of soil procedure will be referred to as

Reagent As and Reagent Bs, and the reagents used in this and other analyses will be referred to as

Reagent Aw and Reagent Bw. Reagent Aw was prepared by combining three different solutions:

12 g of ammonium molybdate dissolved in 200 mL of distilled water, 0.2908 g of antimony

potassium tartrate dissolved in 50 mL of distilled water, and 144 mL concentrated sulfuric acid

(18M) in 1500 mL of distilled water that was allowed to cool. The solution mixture was then

diluted to 2 L, mixed, and stored in a cool, dark place. Prior to each analysis, Reagent Bw was

made by adding 0.75 g of ascorbic acid to 50 mL of Reagent Aw and diluting to 500 mL.

The RPA was determined by placing 10 g of soil in 20 mL scintillation vials and adding 2

mL of 2000 mg L^1 P solution. The vials were capped and vigorously shaken. This was allowed

to equilibrate for 24 hours. Whatman #1 filters were cut to fill the bottom of 50 mL polyethylene




























-* Mihlhapper
--C Taval~es


Table 3-6. Selected characteristics of raw and digested flushed dairy manure (FDM). (n


Total P
mg L^


Dissolved
Reactive P
mg L^
29.5


Dissolved
Organic P
mg L^
1.0


% Total
DRP CODY
mg L^
64 3360

80 1595


Soluble
COD
mg L
1770


Raw FDM
Digested
FDM
Jr COD = C


45.8

48.9


39.1


696 7.3


chemicall oxygen demand


0.5 -

0.4 -



E 0.2 -

10.1

0.10


PareValume


Figure 3-2. Background P of leachate obtained from addition of 50mM KCl solution to soil
columns.









such as FDM, in areas sensitive to P leaching, but if precipitation of P minerals occurs in the

surface horizon, another dynamic process must be considered. Studies have shown that the

concentration of Ca in manure is sufficient to induce precipitation of calcium phosphates (Ca-P),

but the minerals are in non-crystalline form, which makes them more soluble. (Wang et al.,

1995; Hansen and Strawn, 2003). Harris et al. (1994) and Wang et al. (1995), using x-ray

diffraction, were unable to detect any crystalline P minerals in soils that received manure from

dairy operations for many years. Organic acids, Mg, and Si have all been suggested as preventing

crystallization of P minerals (Cooperband and Good, 2002; Harris et al., 1994). Sharpley et al.

(2004) determined ion activity products of manured soils using MINTEQA2 chemical speciation

model and ion concentrations in solutions. They reported tricalcium P and octacalcium P as the

dominant Ca-P forms in manured soils, and hydroxyapatite as the dominant mineral in non-

manured soils. These minerals are not stable in the manured soil and are subject to leaching

(Hansen and Strawn, 2003; Graetz and Nair, 1995.) However, apatite minerals seem to pose

relatively little risk of P leaching in soils affected by manure as long as soil pH remains in the

neutral to alkaline range. Wang et al. (1995) did not find any higher levels ofP in leachates

derived from soils spiked with apatite than non-spiked soils. If, as suspected, Ca-P phases are

precipitating, then knowing the specific form will indicate solubility and release potential.

Hansen and Strawn (2003) indicated Ca-P minerals such as octacalcium P and tricalcium

P controlled the P concentration of the soil solution in manure amended alkaline soils.

Desorption experiments they performed showed a rapid release of P followed by a slow, steady

release. This was done using 1g of soil extracted with 20 mL of 0.005 M NaC1. After 12

replenishments of such extractions, 29% of the total P was desorbed from the surface horizon

and 8% was desorbed from the subsurface horizon. Wang et al. (1995) used simulated rainfall to









found preferential adsorption of PO4 OVer dissolved organic P. Because the number of phosphate

groups on a molecule increases its sorption, and the number and density of phosphate groups in

soil organic matter is variable, the adsorption characteristics vary as well, which explains

differing results when the P source and form changes (Qualls and Haines, 1991; Leytem et al.,

2002).

Leaching of P when using organic and inorganic fertilizers is often studied. Siddique et

al. (2000) conducted leaching trials using five loam soils with P from anaerobically digested

sewage, processed into dry biosolids, or inorganic Ca-P. Both sources of P were amended into

the different soils and placed in soil columns (6.5 cm diameter and length of 30 cm) to a depth of

20cm. Although the biosolids contain organic P, after the digestion process, up to 80% of P is

inorganic, and up to 97% of the P leached was inorganic. The columns were leached with a total

of 5 L of deionized water. In most cases more P was released from the inorganic fertilized soils.

This was attributed to slower P saturation in the biosolid treated soils caused by less soluble P in

the biosolid amendments compared with the Ca-P fertilizer. Jiao et al. (2004) compared loads of

P from soils receiving either organic or inorganic fertilizer. The soil was a fine-silty, mixed,

frigid Typic Endoaquent. The composition of the soil was 300 g kgl sand, 540 g kgl silt, and

160 g kgl clay with 15.4 g total C kgl and pH 6.1 in the 0- to 15-cm layer. Soil columns (10 cm

diameter and 20 cm depth) were leached with synthetic rainwater and nutrient loads were

calculated. Soils receiving inorganic fertilizer had 48% less dissolved reactive P load than soils

receiving organic fertilizer. They stated that the soil had less sorption capacity for dissolved

organic P than for inorganic P. The dissolved reactive load was positively related to the soil

Mehlich-3 P concentration (R2 = 0.50). Carefoot and Whalen (2003) studied leaching of P from a

silt-loam Gleysol fertilized with inorganic (triple superphosphate) and organic (composted cattle












































0l 5 10) 15 20
Pore Volume


25 3035


Figure 3-3. Total P (TP) in leachate for each pore volume. Definite breakthroughs can be seen in
the columns receiving the inorganic P amendments contrasted with the lack of
breakthroughs in the columns receiving the flushed dairy manure (FDM)
amendments. A) Millhopper TP. B) Tavares TP. C) Orsino TP.


C)~f~B~L~PC~sr~m~i~4;~d 48~


YIII-YIU


-

-


58

QO

36
j
E



r,








4Q

'51)
j
m
Em
P


21'


40
3-


;J


O1 5 10 15 20 25 30) 35
Pore Volume


C.


~144j~:~Cll~lllmllUd


0 5 101 16 20
Pore~ Volume


25 3035


--P hosphate
-*-igested FDM
----RawFDM












Soils and P Accumulation: FDM vs Phosphate .....___._ ..... ... .__. ......._.........3


4 DI SCUS SSION ........._.___..... .___ ...............42....


Future Studies .............. ...............48....

Summary and Conclusions .............. ...............49....


LITERATURE CITED .............. ...............49....


BIOGRAPHICAL SKETCH .............. ...............54....









P-M-sand + OM-M & OM-M-P-M-sand,

OM-M-P-M-sand & OM + M-P-M-sand,

M-P-M-sand + P & P-M-P-M-sand,

and so on. The significance of this hypothetical P sink may not be great, but perhaps a study is

warranted.

This experiment indicates that P entering the soil from FDM application may be in

several "pools" that each has its own release constant. About '/ to '/ of the P, from the digested

or raw FDM, respectively, is contained in the particulate fraction, which is subj ect to varying

rates of decomposition as soil humus. The remainder may precipitate with Ca, sorb with Fe and

Al oxides or become completed with OM along with metal ions. In the soils used in this

experiment, the FDM immediately creates conditions that result in less P leaching than occurs

with inorganic P amendments, which effectively rejects the original hypothesis. Indeed, any

organic transport is overridden by the accumulation caused by precipitation and entrapment of P

in OM. The rate of FDM addition and the rate of rainfall will be important factors concerning

future P leaching. Based on this study and others, P, as delivered by FDM, will likely accumulate

to high levels in these soils (Lehmann et al., 2005; Graetz and Nair, 1995). Most of the P will

likely be metastable Ca-P along with P associated with mineralizable organic particulates, both

of which are ultimately subj ect to leaching under rainfall, particularly when FDM (and hence Ca)

additions cease. The complexation of inorganic P with the OM will affect its mobility depending

on soil properties down the profile, which can be very different in originally acidic soils.

Eventually these soils may leach P at slow, steady rate.

The relatively small amount of P leached when using digested or raw FDM was very

similar between these two treatments. However, less P was removed as DRP for the digested









centrifuge tubes that had 6 to 8 small holes in the bottom. The filter was also wetted with 50 mM

KCl and centrifuged by itself at 1500 g for 5 min to drain excess KCl solution. The wet soil was

transferred to the centrifuge tubes and collection cups were taped to the bottom. The tubes were

then centrifuged at 1500 g for 5 min in order to extract the soil solution. The solution was then

filtered through 0.45 Cpm filters, and 0.1 mL aliquots were put in glass centrifuge tubes. A set of

P standards were made by transferring 0.1 mL of 0, 50, 100, 200, 400, 600, 800, 1000, 1500, and

2000 mg L1 P standards in glass centrifuge tubes. All were placed in a drying oven at 70oC

overnight. After drying, 20 mL of 0. 1M HNO3 WAS added to all tubes. The tubes were mixed

vigorously and 0.5 mL aliquots were added to 5 mL Reagent Bw. Color development proceeded

for two hours, at which time absorbance readings were taken at 880 nm.

The amount of P adsorbed (S) was calculated using the formula:

S = [(2000 Cps) x 2] / 10,

where S is the amount of P adsorbed (Cpg/g of soil), Cps is the P concentration in pore solution

after 24 hour equilibration with soil (Cpg/mL). With S, the RPA was calculated thusly:

RPA = S/400,

where 400 Cpg/g represents the maximum adsorption possible under the experimental conditions.

Particle Size Distribution Analysis (PSDA) of Soils

The particle size distribution for each soil was determined (Soil Survey Staff, 1992). To

begin, 100 mL of Calgon (5% solution of sodium hexametaphosphate) was added to 250 mL

Erlenmeyer flasks containing 50 g of dried soil. The flasks were stoppered and placed on a

shaker for 16 hours. The contents were transferred into cylinders and were diluted to 1000 mL

with deionized water while thoroughly rinsing the flasks. The cylinders were shaken and inverted

6 times. They were then placed in a bath at 3 min intervals. After the allotted time, a glass pipette

was lowered into the cylinder at the allotted depth to withdraw the clay fraction. The clay









Bwl of the Levy County typical pedon, with the exception that the matrix color was 10YR 5/6

instead of 10YR 6/4.

Soil Characterization

The results of the analyses characterizing the soils in Table 3-4 show some important

properties for P adsorption and mobility in sandy soils. The TP of the three soils were typical and

indicated little or no impact from anthropogenic activities regarding P loads. The RPA varied

considerably among the soils with the Tavares soil having an unexpectedly high value, but there

was a commensurate increase in the amount of Fe and Al oxide present, and there was high

correlation (r2 = 0.90) between the RPA and the total amount of oxalate extracted Fe and Al

(Figure3-1) The Orsino Bw horizon had the highest percentage of fine sand (Table 3-5),

consistent with its formation in dunal parent material which characteristically is well sorted and

dominated by fine sand. Note that the Tavares El and Millhopper El were not used in the

column experiment.

Flushed Dairy Manure (FDM) Characterization

Mean TP values of the raw and digested FDM are similar whereas the mean amount of

DRP in the digested FDM is about 25% more than in the raw FDM (Table 3-6). The spike

recovery test results showed no interference in the DRP testing procedure. The digestion process

removed over half the COD present in the wastewater, but did not appear to significantly change

the pH.

Analyses of P in Leachate

Background P values (Figure 3-2) stabilized after 6 pore volumes. Less than 0.1 mg L1 P

was detected in any horizon with the Millhopper E2 horizon having slightly higher values. Soil

differed with respect to retardation of P applied as inorganic P (Figure 3-3). The Millhopper soil

started to have breakthroughs of P on the second pore volume. The Orsino and Tavares soils











Rhue, R.D., W.G. Harris, G.Kidder, R.B. Brown, and R. Littell. 1994. A soil-based phosphorus
retention index for animal waste disposal on sandy soil. Final Report, Florida
Dep.Environ. Protection, contract number WM459.

Rhue, R.D., W.G. Harris, V.D. Nair. 2006. A retardation-based model for phosphorus transport
in sandy soils. Soil Sci. 171:293-304.

Sharpley, A.N., R.W. McDowell, and P.J.A. Kleinmann. 2004. Amounts, forms, and solubility of
phosphorus in soils receiving manure. Soil Sci. Soc. Am. J. 68:2048-2057.

Sharpley, A.N., J.L. Weld, D.B. Beegle, P.J.A. Kleinman, W.J. Gburek, P.A. Moore, and G.
Mullins. 2003. Development of phosphorus indices for nutrient management planning
strategies in the United States. J. Soil Water Conserv. 58:137-152.

Siddique, M.T., J.S. Robinson, and B.J. Alloway. 2000. Phosphorus reactions and leaching
potential in soils amended with sewage sludge. J. Environ. Qual. 29: 193 1-193 8.

Slabaugh, J.D., A.O. Jones, W.E. Puckett, and J.N. Schuster. 1996. Soil Survey of Levy County,
FL. USDA-Natural Resources Conservation Service, National Cooperative Soil Survey.

Soil Survey Staff. 1992. Soil Survey Laboratory Methods Manual. p. 14. U.S. Department of
Agriculture.

van Es, H. M., R.R. Schindelbreck, and W.E. Jokela. 2004. Effect of manure application timing,
crop, and soil type on phosphorus leaching. J. Environ. Qual. 33:1070-1080.

von Wandruszka, R. 2006. Phosphorus retention in calcareous soils and the effects of organic
matter on its mobility. Geochem. Trans. 7: Art. No. 6.

Wang, H.D., W.G. Harris, K.R. Reddy, and E.G. Flaig. 1995. Stability of phosphorus forms in
dairy-impacted soils under simulated leaching. Ecol. Eng. 5:209-227.

Wilkie, A.C. 2005. Anaerobic digestion: biology and benefits. In: Dairy Manure Management:
Treatment, Handling, and Community Relations. NRAES-176:63-72.

Wilkie, A.C. 2005. Anaerobic digestion of dairy manure: design and process considerations. In:
Dairy Manure Management: Treatment, Handling, and Community Relations. NRAES-
176:301-312.

Wilkie, A.C., H.F. Castro, K.R. Cubinski, J.M. Owens, and S.C. Yan. 2004. Fixed-film
anaerobic digestion of flushed dairy manure after primary treatment: wastewater
production and characterization. Biosystems Engineering 89:457-471.

Zhou, M., R.D. Rhue, and W.G. Harris. 1997. Phosphorus sorption characteristics of Bh and Bt
horizons from sandy coastal plain soils. Soil Sci. Soc. Am. J. 61:1364-1369.









ACKNOWLEDGMENTS

I thank Dr. Willie Harris and Dr. Ann Wilkie for all their help and mentorship. I appreciate

the accessibility and helpful technical advice provided by my other committee members, Dr.

Dean Rhue and Dr. Vimala Nair. I also thank my wife, Elizabeth, for her unwavering support.













7-
6-




3-

3-

2-
1-


Orsino P Tavrares P Millhopper P Orsino D


T~avares D Millhopper D Millhop~per R


Figure 3-6. P accumulated in soils receiving different P amendments. "P", "D", and "R" indicate
inorganic P, digested flushed dairy manure (FDM), and raw FDM amendments,
respectively. Orsino R and Tavares R values are absent due to column clogging
resulting in discontinued P additions. Error bars represent standard deviation.


Table 3-7. ANOVA output
Df
P source 2
Soil 2
P source soil 4
Rep 2


Type III SS
35.34007
8.57565
10.88270
0.04502


Mean Square
17.67003
4.28782
2.72067
0.02250


F value
747.60
181.41
115.11
0.95


Pr > F
< 0.0001
< 0.0001
< 0.0001
0.4095


Table 3- 8. Duncan's Multiple Range Test for Accumulated P. The mean accumulated
inorganic P was significantly different than the mean accumulated digested
flushed dairy manure (FDM) and raw FDM. The mean accumulated P in the
FDM treatments were not significantly different. Clogging of soil columns
receiving raw FDM resulted in missing data for two soil columns.
Duncan grouping Mean (mg P) N P form
A 6.0 9 Digested FDM
A 5.5 7 Raw FDM
B 3.3 9 KH2PO4









manure) P sources by measuring P in the subsurface water at a 60cm depth. It contained 0.3 to

1.7 mg P L Particulate P was the dominant P form at most sampling dates. Also, Phillips

(2002) studied leaching of P from piggery wastewater using a Vertisol and a Spodosol in

undisturbed soil cores (diameter of 30 cm and depth of 60 cm). Both molybdate reactive P and

unreactive P were leached. The author lists unreactive P as dissolved organic P, particulate P,

and non-reactive P. The Spodosol leached mostly molybdate reactive P (approximately 70%)

because the wastewater contained about 70% of this form, and the soil had little adsorption

capacity. The P leached from the Vertisol was mostly (approximately 80%) unreactive P because

the molybdate reactive P was adsorbed by the soil colloids. These studies highlight differences in

leaching potential of P from soils when different forms of inorganic and organic P are applied to

different soils.

The interactions of organic matter with P and P adsorption sites are a critical area of

research concerning P mobility. Organic colloids have been shown to transport metal

contaminants in subsurface soils. Karathanasis and Johnson (2006) reported higher metal

elutions up to four orders of magnitudes greater than the controls. Biosolids, derived from

municipal wastes, and poultry manure were applied to an Alfisol, a Mollisol, and an Entisol. The

metals, Cd, Mo, and Cr, were present in both the particulate and soluble fractions, and the

significant increase was attributed to increased organic-metal complexation and exclusion

facilitators. lyamuremye et al. (1996) found that organic amendments to soil decreased P

sorption sites and sorption that was related to changes in pH and exchangeable Al. Chardon et al.

(1997) conducted a set of experiments examining the leaching potential of dissolved organic P

(DOP) in cattle slurry applied on sandy soil. The DOP in the slurry consisted mainly of high

molecular weight compounds. Leaching of P was examined using laboratory soil columns









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

ENHANCED RETENTION OF PHOSPHORUS APPLIED AS FLUSHED DAIRY MANURE


By

Aaron Malek

December 2007

Chair: Name Willie Harris
Cochair: Ann Wilkie
Major: Interdisciplinary Ecology

Leaching of phosphorus (P) is a maj or concern in many areas of the world, especially in

environments with karst topography that contain sensitive and valuable aquifers and springs. In

one such area, North Florida of the SE United States, dairy farming is prevalent, and in such

operations it is common to spray crop fields with flushed dairy manure (FDM) as a means of

nutrient recycling. We investigated the fate of P in three sandy soils that typify and encompass

large areas of the region. At the University of Florida Dairy Research Center, anaerobic digestion

of FDM is currently practiced and provides many benefits including reduction of organic matter

(OM) and mineralization of nutrients, including P. Three different P amendments were applied to

soil columns in a randomized complete block design: raw FDM, digested FDM, and inorganic P.

Since OM is known to form complexes and compete for sorption sites with P, thereby reducing P

sorption, it was hypothesized that retardation of P would be least in soils receiving P as raw

FDM and most in soils receiving inorganic P. Soil columns were kept under 20 cm of suction to

promote unsaturated flow and amendments were applied at one half of the soil pore volume.

Leachate was collected per pore volume and analyzed colorimetrically for total P (TP) and

dissolved reactive P (DRP). Little evidence of OM transport of P was found in columns receiving









leach dairy manure impacted soil samples in soil columns. After 17 weeks of simulated rainfall:

each week receiving synthetic rain 80% of the soil pore volume, 13% to 26% of the total P was

leached from the most impacted areas. Graetz and Nair (1995), after 10 sequential extractions on

A and Bh horizons of soils from active dairies found 2% to 18% of the total TP was extracted.

The soils, however, kept releasing P at diminished levels in the concluding extractions. They also

found higher SRP concentrations in Bh horizons of abandoned dairy sites compared to active

sites (43 mg kg-l to 14 mg kg- respectively), supporting the notion that P, over time, will

continue to be released and leach. It should also be noted, however, that these sandy soils had

little to no P retention capacity, especially the A and E horizons, unlike the Tavares soil used in

this experiment.

The change that manure amendments, such as FDM, have on soils is remarkable with

regards to P fate and transport. As demonstrated in this experiment, P delivered as KH2PO4

solution behaved very differently than P delivered as either digested or undigested FDM in the

same soils. The leaching and accumulation of P from inorganic P additions could seemingly be

predicted from the RPA values of the soils (Figure 3-7). The reported change manure

amendments bring about in acid soils, such as these, is a shift from Al and Fe reaction products

to Ca/Mg reaction products, due largely to the great influx of manure-derived bases increasing

soil pH (Harris et al. 1994; Sharpley et al. 2004). This change is largely confined to the surface

horizons as Graetz and Nair (1995) show the shift from Ca/Mg-P predominance in the surface

horizon to Al/Fe-P predominance in the deeper horizons and similar nonimpacted soils have

Al/Fe-P predominance throughout the profile. This is a significant phenomenon because the

factors controlling P mobility change as P leaches through the soil. For example, the most

important factor organic matter may play in the surface horizon is the inhibition of stable Ca-P









aliquots were placed in weighed, metal cups, dried in an oven, and weighed again the following

day. The contents of the cylinders were then rinsed through a No. 325 sieve (48-Clm mesh) to

remove the clay and silt. The remaining sand was collected and after drying, it was fractionated

into very coarse (2000-1000 Clm), coarse (100-500 Clm), medium (500-250 Clm), fine (250-100

Clm), and very fine (100-48 Clm particles using sieve Nos. 18, 35, 60, 140, and 325. Sand

fractions were weighed and the distribution was calculated using the formulas:

%Sand = (sand weight) (100) / (sample weight),

%Clay = { [(metal cup + clay weight + Calgon) metal cup blank weight](40)}(100) / (sample

weight), %Silt = 100% %Clay %Sand.

Flushed Dairy Manure Analyses

Flushed dairy manure samples were collected at the Dairy Research Unit in Hague,

Florida. The raw FDM was collected from the pump sump that feeds into the digester. The

digested FDM was obtained directly from the effluent line. All samples were stored in a cold

room at 40C. The FDM samples were primarily analyzed for TP, DRP, pH, and chemical oxygen

demand (COD). The samples from September 14, 2005 were chosen as treatments for this study

out of all samples taken because the digested and raw FDM showed typical differences in DRP

and COD, i.e., DRP was greater, and COD less, for digested FDM. It was important for the

purposes of this study that digested and raw FDM show these differences because the

hypothesized differences in potential P leaching relates to the proportion of DRP to TP. The pH

was measured with a standard, portable pH meter. The TP was determined by taking a 5 mL

aliquot of wastewater and diluting to 10 mL with deionized water. From this mixture, 0. 1 mL

was taken and placed in a Hach COD reactor test tube. This was done so no more than 2.5 Cpg of

P would be contained in each test tube. The samples were dried and then ashed in a muffle

furnace at 300oC for 1 hour and at 550oC for two more hours. Aliquots of 0. 1 mL P standard









(diameter of 15.3 cm and length of 100 cm) and outdoor lysimeters. The soil columns contained

a quartzitic sand, with a total P of 6.5 mg kg- up to 70 cm with a 16 cm layer of a sandy loam,

containing 530 mg kg- total P placed on top. The columns were leached with 100 mL of water

twice a week for four months. Of the total P leached, more than 90% from the soil columns and

more than 70% from the lysimeters leached as DOP, indicating its high mobility. Also, DOP in

the soil pore water increased from about 10% of total P in the topsoil to more than 70% at 70 to

80 cm depth. Because the leaching of total P mainly occurred in periods of low Cl and NO3

concentrations, indicative of high leaching rates, the authors suggested that the P transport was

mediated by dissolved organic carbon and other colloidal particles. Eghball et al. (1996) also

found deeper movement of P in soils receiving manure application than soils receiving equal

amounts of inorganic P. They, too, suggested that the P moved in organic forms or chemically

reacted with compounds in manure, enhancing solubility. This was examined further by

Motoshita et al. (2003). They studied leaching of colloidal and dissolved P in soil column

experiments using a surface loam soil with a high Olsen-P content (93 mg-P/kg). The columns

were leached with artificial irrigation solution. Colloidal P leaching showed a minor increase

with time, and dissolved P leaching was nearly constant. Dissolved P consisted of 81-86% of

total P leached. There was a high correlation between dissolved P leaching and dissolved organic

matter leaching (R2 = 0.82). The P sorption was investigated and showed that the P was sorbed to

or formed complexes with the dissolved organic matter. These studies show the importance of

organic forms of P and dissolved organic matter relating to P mobility in the soil.

In addition to the interactions affecting P mobility, many have investigated the physical

pathway and flow that P takes within the soil. This is especially important for finer textured

soils. Dj odjic et al. (2004) sought to establish a relationship between soil P levels and actual P









BIOGRAPHICAL SKETCH

Aaron Malek was born in Bryan, TX. He graduated from Navasota High School in 1998

and then received a Bachelor of Science in Interdisciplinary Biology from the University of

Florida in 2004. He enrolled in graduate school, also at the University of Florida, the same year

at the School of Natural Resources and the Environment. He will graduate in December 2007

and receive a Master of Science in Interdisciplinary Ecology.









that P movement varies from soil to soil (Dodjic et al., 2004; Elrashidi et al., 2001; van Es et al.,

2004) Many properties of soils have been shown to have an impact on P behavior. The

complexity resulting from the multiple factors determining P behavior in soils has made it the

subject of many studies. This review will examine literature concerning P movement when

delivered in different forms, such as organic or inorganic fertilizers, the characteristics of

different forms of P and its sources, especially raw and digested FDM, and the interactions of P

in the soil that relates to its retention and leaching potential.

As the delivery and specific characteristics of P sources are important for understanding

P fate, attention must be given to how it is derived, especially when considering organic

amendments. The production of FDM is one option used by dairy operations to properly deal

with livestock manure. Wilkie et al. (2004) describe a system utilized at the Dairy Research Unit

at the University of Florida in which the FDM is anaerobically digested in a Eixed-film digester.

Dairy facilities are kept sanitary by flushing the manure down narrow alleys into a series of

storage ponds where separation of bedding sand and very fibrous materials occurs mechanically

and by sedimentation (Figure 1-1). The Eixed-film digester contains media on which microbes

colonize. The microbes include a symbiotic group of anaerobic bacteria which convert complex

organic matter into smaller molecules including methane and carbon dioxide (Wilkie 2005). The

benefits of anaerobic digestion include production of energy-yielding methane, mineralization of

nutrients, including P, reduction of odors, inactivation of weed seeds, and lower pathogen levels.

The reduction and change of the organic matter and mineralization of nutrients may have an

effect on P mobilization when applied to the soil.

Dissolved P interacts with Fe, Al, Mn, and Ca in soils (Brady and Weil, 2002; Pardo et

al., 2003). Many authors have found the amount of Fe and Al in soils to be important in









solutions of 0, 5, 10, 15, 20, 25, and 30 mg L^1 P were transferred into test tubes and were

evaporated. Then, 5 mL of Reagent Bw was added to each sample and standard, and blue color

developed for two hours at which time absorbance readings were taken at 880 nm.

Total P (TP) Method Investigation HCI Digestion vs Only Ashing

Five mL of FDM was diluted to 10 mL with deionized water. After mixing, 0. 1 mL

aliquots, totaling 12, were transferred to Hach COD reactor test tubes. All test tubes were dried

using the COD reactor at 120oC. After drying, all test tubes were placed in a muffle furnace at

300oC for 1 hour and 550oC for two hours. After cooling, 6 of the wastewater samples, two of the

30 mg L^1 P standards, and one of the 0 mg L^1 P standard received 5 mL 6 M HC1. These were

heated in the COD reactor for several hours at 120oC, just above refluxing temperature but, no

visible signs of boiling, until all liquid had evaporated. After cooling, 5 mL of Reagent Bw was

added to each sample and a set of standards, and blue color developed for 2 hours at which time

absorbance readings were taken at 880 nm. The P concentrations of the samples that were HCI

digested with 6 M HCI were compared with those that were not. It was determined that this

modified procedure was satisfactory for the purposes of this study, enabling rapid assessment of

TP in small volumes of column effluent (described below). The mean TP of the FDM that had

HCI digestion was 46.51 0.6 mg L^1 and the mean TP of the FDM with no HCI digestion was

48.6 & 0.8 mg L^1

Dissolved Reactive Phosphorus (DRP)

In order to determine if any constituent of the wastewater interfered with analysis

reagents and skewed results, a spike recovery test was performed. The raw and digested FDM

was filtered through 0.45 Clm filter, and 0.5 mL was diluted to 15 mL with distilled water. From

this, 1 mL aliquots were transferred into test tubes. In duplicate, samples were spiked with 0.1

mL of 0, 5, 10, and 15 mg L^1 P standards. A set of standards was made: 0. 1 mL of 0, 5, 10, 15,









CHAPTER 3
RESULTS

Soil Descriptions

The Tavares sand was collected on May 18, 2005 in Alachua County, FL 604m south of

Lake Mize in the Austin Cary Memorial Forest. The soil was collected by boring three holes, 20

cm apart, with an auger. Tavares sand is derived of sandy marine deposits and is moderately well

drained. It occurs in a forested environment on a <2% slope dominated by oak and long-leaf

pine. The Tavares soil horizons are described in Table 3-1.

The Millhopper soil was collected on May 18, 2005 in Alachua County, FL. University

of Florida, 126 m WNW (3070) from the NW corner of Museum Rd and Newell Dr. The soil was

similarly collected by boring three holes, 20 cm apart, with an auger. It is a moderately well

drained soil derived from sandy marine deposits. It occurs in a small forested area on a <5%

slope dominated by sweetgum, hickory, oak, and other small trees and shrubs. Table 3-2 shows

the descriptions of the Millhopper soil horizons.

The Orsino soil Bw horizon was obtained from the Soil Mineralogy lab at the University

of Florida. It was collected by Willie Harris from a vertical exposure of an old sand dune near

Cedar Key (Levy County), FL, during a professional field trip. It occurred in an Orsino fine sand

map unit and the profile conformed to the Orsino series (personal communication, Willie Harris).

Orsino soils typically occur on elevated ridges and knolls. They have low available water

capacity and fertility, but still support natural vegetation including slash pine, sand pine and

scrub oak. A description of this soil taken from the Levy County Soil Survey (Slabaugh et al.,

1996) is included in Table 3-3. The typical pedon is listed as being located 3,300 feet

(approximately 1000 m) north and 250 feet east (approximately 76 m) of the southwest corner of

sec. 11, T. 14 S., R. 16E. The Bw horizon used in this study was morphologically similar to the










Eghball, B., G.D. Binford, and D.D. Baltensperger. 1996. Phosphorus movement and adsorption
in a soil receiving long-term manure and fertilizer application. J. Environ. Qual. 25:
1339-1343.

Elrashidi, M.A., A.K. Alva, Y.F. Huang, D.V. Calvert, T.A. Obreza, and Z.L. He. 2001.
Accumulation and downward transport of phosphorus in Florida soils and relationship to
water quality. Comm. Soil Sci. Plant Anal. 32:3099-3119.

Fernandez-Perez, M., F. Flores-Cespedes, E. Gonzaalez-Pradas, M.D. Urena-Amate, M.
Villafranca-Sanchez, M. Socias-Viciana, and S. Perez-Garcia. 2005. Effect of dissolved
organic carbon on phosphate retention on two calcareous soils. J. Agr. Food Chem. 53:
84-89.

Freese, D., S.M. Vanderzee, and W.H. Vanriemsdijk. 1992. Comparison of different models for
phosphate sorption as a function of the iron and aluminum-oxides of soils. J. Soil Sci.
43:720-738.

Geohring, L.D., O.V. McHugh, M.T. Walter, T.S. Steenhuis, M.S. Akhtar, and M.F. Walter.
2001. Phosphorus transport into subsurface drains by macropores after manure
applications: Implications for best manure management practices. Soil Sci. 166:896-909.

Graetz. D.A., and V.D. Nair. 1995. Fate of phosphorus in Florida spodosols contaminated with
cattle manure. Ecol. Eng. 5:163-181.

Hansen, J.C., T.C. Daniel, A.N. Sharpley, and J.L. Lemunyon. 2002. The fate and transport of
phosphorus in agricultural systems. J. Soil Water Conserv. 57:408-417.

Hansen, J.C. and D.G. Strawn. 2003. Kinetics of phosphorus release from manure-amended
alkaline soil. Soil Sci. 168:869-879.

Harris, W.G., H.D. Wang, and K.R. Reddy. 1994. Dairy manure influence on soil and sediment
composition: implications for phosphorus retention. J. Environ. Qual. 23:1071-1081.

Hens, M. and R. Merckx. 2001. Functional characterization of colloidal phosphorus species in
the soil solution of sandy soils. Environ. Sci. Technol. 35:493-500.

Hesketh, N., P.C. Brookes, and T.M. Addiscott. 2001. Effect of suspended soil material and pig
slurry on the facilitated transport of pesticides, phosphate, and bromide in sandy soil. Eur.
J. Soil Sci. 52:287-296.

lyamuremye, F., R.P. Dick, and J. Baham. 1996. Organic amendments and phosphorus
dynamics. 2. Distribution of soil phosphorus fractions. Soil Sci. 161:436-443.

Jiao, Y., W.H. Hendershot, and J.K. Whalen. 2004. Agricultural practices influence dissolved
nutrients leaching through intact soil cores. Soil Sci. Soc. Am. J. 68:2058-2068.










LIST OF FIGURES


Figure page

1-1 Schematic layout of University of Florida Dairy Research Unit manure-handling
system; heavy arrows indicate flow of flushwater and manure solids. .........._.... .............19

2-1 Columns containing 20g of soil were kept under 20 cm of suction utilizing
Erlenmeyer flasks to ensure unsaturated flow of leachate, which was collected in
glass vials. .............. ...............29....

3-1 Correlation between total oxalate extracted Fe and Al in molar concentrations and
RPA values of the soils. ........... ..... ._ ...............35..

3-2 Background P of leachate obtained from addition of 50mM KCl solution to soil
colum ns. .............. ...............36....

3-3 Total P in leachate for each pore volume............... ...............37.

3-4 Dissolved reactive P (DRP) concentrations in leachate for each pore volume..................38

3-5 Phosphorus in leachates for flushed dairy manure (FDM) treatments, shown as both
total P and dissolved reactive P (DRP). ............. ...............39.....

3-6 Phosphorus accumulated in soils receiving different P amendments. ............. ..............40

3-7 Correlation between P accumulated in soil columns, delivered as inorganic P at 48
mg L1 P solution, and soil RPA values. ............. ...............41.....









sample, 2 g of soil was placed in 50 mL centrifuge tubes that were wrapped in aluminum foil to

exclude light. A 50-mL aliquot of acid oxalate extraction solution was added to the soil in tubes,

which were then capped tightly and placed in a reciprocal shaker at low speed for 4 hours. Tubes

were centrifuged for 20 min at 2000 rpm, and supernatant solutions filtered through Whatman 42

filters and stored cold (4oC) until analysis within a few days.

An atomic absorption spectrophotometer was used to determine Fe and Al content. The

standard solutions were prepared so that the matrix contained the same concentration of acid

oxalate as the extracting solutions.

Total P (TP) of Soils

The TP of the three soils was determined by using an ashing and acid digestion procedure

(Anderson, 1976). A 50 mL beaker containing 1 g of soil was placed into a muffle furnace at

3500C for one hour and at 5500C for two more hours. After cooling, 20 mL of 6M HCI was

added, and it was allowed to slowly evaporate on a hotplate. After the sample dried, the

temperature was raised briefly and then allowed to cool. Then 2.2 mL of 6M HCI was added, and

the beakers were heated so that the residue was easily dislodged. The mixture was then

transferred to a funnel with a Whatman #42 filter paper, and the solution filtered into 50 mL

volumetric flask. The beakers and the filters were rinsed repeatedly with deionized water, and the

volume of the flasks was brought up to 50 mL.

Analysis of P in extracts was done colorimetrically (Murphy and Riley, 1962), with P

standards ranging from 0 to 35 mg L 1. 0.5 mL of each standard and unknown solution was

placed in a test tube. Reagent As was prepared by dissolving 12 g of ammonium molybdate in

100 mL of H20 and 0.2908 g of antimony potassium tartrate in another 100 mL of H20; both of

these solutions were added to 1 L of 2.5M H2SO4 and diluted to 2 L. This was kept stored in a










LIST OF TABLES


Table page

3-1 Tavares soil description .............. ...............33....

3-2 Millhopper soil description ................. ...............33................

3-3 Orsino typical pedon description, taken from the Levy County Soil Survey Report ........34

3-4 Characterization of soils collected. (n = 3) .............. ...............35....

3-5 Sand size distribution among soils collected and pore volume of soil horizons used in
the experiment ................. ...............35........._.....

3-6 Selected characteristics of raw and digested flushed dairy manure (FDM). (n = 3) .........36

3-7 ANOVA output ................. ...............40........... ....

3- 8 Duncan's Multiple Range Test for Accumulated P. ............. ...............40.....









20, 25, and 30 mg L^1 P. Five mL of Reagent Bw was added to each sample and standard, and

blue color developed for two hours at which time absorbance readings were taken at 880 nm.

The increasing spike was checked to be linear (r2 > 0.998).

Chemical Oxygen Demand (COD)

The chemical oxygen demand (COD) of the raw and digested FDM was determined as an

indication of digestion. The mg L1 COD results are defined as the mg of 02 COnSumed per liter

of sample used in the procedure. The anaerobic digestion process eliminates a significant amount

of the COD from the raw FDM. COD analyses were performed as described in Wilkie (2004).

First, 500 mL of each FDM was homogenized in a blender for 2 minutes, and a 2 mL aliquot was

added to the COD Digestion Reagent Vial provided by Hach. Also, a subsample of the

homogenized FDM was centrifuged at 12,000 min' for 30 min, and the supernantant was

sampled to determine the soluble COD. Vials were heated in the Hach COD reactor for 2 hours

at 150oC. The reagent contains dichromate (Cr207-2), which oxidizes organic compounds and is

reduced to the green chromic ion (Cr+3). The amount of Cr+3 prOduced is determined

colorimetrically from which the COD is calculated.

Soil Column Experimental Procedures

Column Construction

Twenty seven cylindrical soil columns were used in this experiment. They have an inside

diameter of 1.6 cm and stand 20 cm tall. The columns have a bushing-like fitting that contains a

glass frit at the bottom. A nipple at the bottom and a piece of glass tubing are made continuous

by a piece of clear plastic tubing. The collection vial encompasses the glass tubing so that

suction can be applied to the vial, pulling the soil leachate under unsaturated flow. The

continuous suction is supplied by an elevated Erlenmeyer flask that has a siphon held to it. Air

leaks anywhere on the column are mediated by tape and petroleum sealant. The elevation of the










LITERATURE CITED


Agbenin, J.O. and S.O. Igbokwe. 2006. Effect of soil-dung manure incubation on the solubililty
and retention of applied phosphate by a weathered tropical semi-arid soil. Geoderma 133:
191-203.

Anderson, J.M. 1976. An ignition method for determination of total phosphorus in lake
sediments. Water Res. 10:329-331.

Andrade, F.V., E.S. Mendonca, V.H.V. Alvarez, and R.F. Novais. 2003. Addition of organic and
humic acids to latosols and phosphate adsorption effects. Rev. Bras. Cienc. Solo 27:
1003-1011.

Akhatar, M.S., B.K Richards, P.A. Medrano, M. DeGroot, and T.S. Steenhuis. 2003. Dissolved
phosphorus from undisturbed soil cores: Relation to adsorption strength, flow rate, or soil
structure? Soil Sci. Soc. Am. J. 67:458-470.

Baril, R. and Bitton, G. 1967. Anomalous values of free iron in some Quebec soils containing
magnetite. Can. J. Soil Sci. 47:261.

Bennani, F., M. Badraoui, M. Mikou. 2005. Monocalcium phosphate monohydrate concentration
in soil suspension amended with organic matter. J. Phys. IV 123:159-163.

Brady, Nyle C. and R. Weil. The Nature and Properties of Soils. 13th Ed. Upper Saddle River,
New Jersey: 2002.

Carefoot, J.P. and J.K. Whalen. Phosphorus concentrations in subsurface water as influenced by
cropping systems and fertilizer sources. Can. J. Soil Sci. 83:203-212.

Chardon, W.J., O. Oenema, P. delCastillo, R. Vriesema, J. Japenga, and D. Blaauw. 1997.
Organic phosphorus in solutions and leachates from soils treated with animal slurries. J.
Environ. Qual. 26:372-378.

Cong, P.T. and R. Merckx. 2005. Improving phosphorus availability in two upland soils of
Vietnam using Tithonia diversifolia H. Plant Soil 269: 11-23.

Cooperband, L.R. and L.W. Good. 2002. Biogenic phosphate minerals in manure: implications
for phosphorus loss to surface waters. Environ. Sci. Technol. 36:5075-5082.

Delgado, A., A. Madrid, S. Kassem, L. Andreu, and M.D. del Campillo. 2002. Phosphorus
fertilizer recovery from calcareous soils amended with humic and fulvic acids. Plant Soil
245:277-286.

Dodjic, F., K. Borling, and L. Bergstrom. 2004. Phosphorus leaching in relation to soil type and
soil phosphorus content. J. Environ. Qual. 33:678-684.









adsorbed virtually all P until the 9th and 19th pOTO VOlume, respectively. This order of P

adsorption capacity is in agreement with the RPA values of the soil horizons.

In all soils, there were higher values of TP in the leachate for the first few pore volumes.

These values decreased until the 5th pOTO VOlume. The largest difference of TP between raw and

digested FDM amended soils was in the Millhopper soil where leachate from the digested FDM

was slightly higher than those derived from the raw FDM. The leachate of the Tavares soil

derived from the raw and digested FDM showed little differences in TP values. There were also

little differences from the Orsino soil, except for the leachate from the raw FDM having higher

values initially and the digested FDM leachate being higher at the last pore volumes. In all cases,

after the breakthrough of P derived from the inorganic P treatment occurred, the level of TP in

the leachate derived from the raw and digested FDM was lower than those derived from

inorganic P.

The DRP in the leachate roughly follows the same trends as the TP (Figures 3-3, 3-4, and

3-5). Leachates derived from the raw and digested FDM had low levels of DRP for the Tavares

and Orsino soils, except for an increase in the last pore volumes derived from the digested FDM

in the Orsino soil. The DRP in the leachate from the raw and digested FDM from the Millhopper

soil is higher than other soils. There is an increase in the last pore volumes as well as a spike in

concentration at the 20th pOTO VOlume. The largest differences between TP and DRP values occur

during the initial pore volumes (Figure 3-5). Afterwards the levels of P are more closely aligned,

especially for Tavares and Orsino.

Soils and P Accumulation: FDM vs Phosphate

Calculated P accumulation in the soils (Figure 3-6) shows that the raw and digested FDM

accumulate similarly in each soil and among all soils, and to a significantly greater extent than

for the inorganic P treatment (Tables 3-7 and 3-8). These values were derived by subtraction of P









CHAPTER 2
MATERIALS AND METHODS

Soil Selection and Sampling

Sandy horizons were sampled from soils of three series that occur in the coastal plain of

the southeastern USA: Tavares (sandy, hyperthermic, Typic Quartzipsamment), Millhopper

(loamy, siliceous, hyperthermic, Grossarenic Paleudult), and Orsino (sandy, hyperthermic,

Spodic Quartzipsamment). Tavares and Millhopper soils are extensive in Florida across a range

of moderately-well-drained landscapes, as are the taxonomic families they represent. Orsino soils

are commonly associated with old dunes or other landforms underlain by deep sandy parent

materials and occupied by oak scrub or sand pine plant communities. These three soils

encompass a wide range of sandy soil horizons and environments in Florida, and were selected

on this basis. Other research criteria satisfied by these soils is that they (i) prospectively provide

an intermediate and representative range of P retardation and (ii) they occur on some leaching-

prone landscapes which are the types of soil/landscapes to which results of this study should

apply. The E2 horizons of the Millhopper and Tavares and the Bw horizon of the Orsino were

chosen for the column experiment described below. All soils were air dried and sieved through a

2 mm sieve prior to experimental use.

Soil Analyses

Oxalate-Extractable Iron (Fe) and Aluminum (Al)

The Fe and Al content of the soils was determined by extraction with acid ammonium

oxalate in the dark (Baril and Bitton, 1967; McKeague and Day 1966). This method extracts both

amorphous inorganic Fe and Al and organic completed Fe and Al in soils. The acid oxalate

extracting solution was made by mixing 700 mL of 0.2 M ammonium oxalate and 535 mL of 0.2

M oxalic acid together and adjusting pH to 3.0 by using either of the base solutions. For each









determine if any crystalline precipitant formed, and leaching of the soil columns would be very

useful in determining the stability of the accumulated P. Also, another column study involving

soils, with a range of P sorption capacity (such as those used in this experiment), overlaid with a

highly impacted soil of similar soil type with large amounts of accumulated P may show how P

in the leachate is affected by changes in the soil chemistry and if OM leaches into the subsurface

layer and prevents sorption at Al/Fe sorption sites.

Summary and Conclusions

Three different sandy soils in a column experiment showed processes related to P

leaching differ when delivered as inorganic P as opposed to when delivered by both raw and

digested FDM. The capacity of the soils to retard the movement of P when amended with

inorganic P is related to the amount of Fe and Al oxide present in the soil. In contrast, the soils

amended with FDM accumulated P and very little leached through the columns. After the

addition of 30 pore volumes, the amount of P leached from the columns amended with inorganic

P was significantly greater than those that received both raw and digested FDM. It is apparent

that FDM creates conditions in the soil that promote accumulation of P. This is likely due to an

increase in pH, from the addition of base cations present in the FDM, notably Ca, and a shift

from Al/Fe-P to Ca/Mg-P reaction products. Also, any transport of P via OM is minimal

compared to the factors promoting P accumulation, and the immobility of OM may also be

caused by the changes FDM creates in the soil. The formation of meta-stable Ca-P should be

verified by analytical means, and leaching of the columns would provide data on the stability of

the P accumulated in all of the soil columns. Also, as there is a shift back to the original soil

chemistry in deeper horizons, leaching of FDM amended soils over the same or similar

nonimpacted soils in a column experiment could help determine what factors are important with

relation to P leaching as the soil chemistry changes.









minerals, whereas in the deeper horizons its role of preventing stable Al/Fe-P sorption may be

most important. Also, the role of changing pH is important. In the surface horizon the increased

pH allows the precipitation of Ca-P minerals leading to their, at least transient, accumulation. In

the more acidic subsurface layers, precipitation is not favored. However Harris et al. (1994)

found an apatite-like mineral in stream sediments associated with, but not in, soils that received

the manure directly. This may indicate that, in some cases, elements near the P source may

inhibit crystallization of apatite, but elsewhere precipitation of this stable Ca-P may occur.

The lack P breakthrough when delivered as FDM shows that, contrary to the original

hypotheses, retardation of P movement was not reduced by competition of FDM organic

components for P adsorption sites and that mediated transfer of P by dissolved organic matter

(OM) in the FDM does not play an immediate role in P leaching. Most of the organic P in the

FDM is predominant in a particulate form, which is prone to entrapment, which would minimize

leaching in the short-term, rather than in dissolved form, which could favor reduced adsorption

to soil components and greater leaching. The microbes filtered out of the FDM by the sand in the

columns would almost certainly flourish in the column environment. Their growth could retain P

and trap it in cellular material. The extent, capacity, and short and long term significance of this

activity, given the large concentration of P being added, is unknown, but as evident in the

difference in COD between raw and digested FDM shown in Table 3-6, the food to

microorganism ratio would be higher for the raw FDM. This may also play a factor concerning

leaching when FDM is used as fertilizer for crops.

The role of organic matter (OM) colloids has been reported as a dominant process in P

mobility in high P soils (Motoshita et al., 2003). Hens and Merckx (2001) and others have

shown that organic matter-metal-orthophosphate (OM-M-P) complexes form in soils. The OM




Full Text

PAGE 1

ENHANCED RETENTION OF PHOSPHORUS APPLIED AS FLUSHED DAIRY MANURE By AARON MALEK A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2007 1

PAGE 2

2007 Aaron Malek 2

PAGE 3

To my Mom 3

PAGE 4

ACKNOWLEDGMENTS I thank Dr. Willie Harris and Dr. Ann Wilkie for all their help and mentorship. I appreciate the accessibility and helpful technical advice provided by my other committee members, Dr. Dean Rhue and Dr. Vimala Nair. I also thank my wife, Elizabeth, for her unwavering support. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES.........................................................................................................................8 ABSTRACT.....................................................................................................................................9 CHAPTER 1 INTRODUCTION................................................................................................................. .11 Background.............................................................................................................................11 Hypotheses..............................................................................................................................12 Objectives...............................................................................................................................12 Literature Review.............................................................................................................. .....12 2 MATERIALS AND METHODS...........................................................................................20 Soil Selection and Sampling...................................................................................................20 Soil Analyses..........................................................................................................................20 Oxalate-Extractable Iron (Fe) and Alum inum (Al).......................................................20 Total P (TP) of Soils........................................................................................................21 Relative Phosphorus Retent ion Capacity (RPA).............................................................22 Particle Size Distribution Analysis (PSDA) of Soils......................................................23 Flushed Dairy Manure Analyses............................................................................................24 Total P Method Investigation HC l Digestion vs Only Ashing.....................................25 Dissolved Reactive Phosphorus (DRP)...........................................................................25 Chemical Oxygen Demand (COD).................................................................................26 Soil Column Experimental Procedures...................................................................................26 Column Cons truction.......................................................................................................26 Pore Volume Determination............................................................................................27 Background P..................................................................................................................27 Flusded Dairy Manure (FDM ) Storage and Sampling....................................................27 Leaching Procedure.........................................................................................................28 Leachate Analyses...........................................................................................................28 Statistical Analyses..........................................................................................................2 8 3 RESULTS...................................................................................................................... .........30 Soil Descriptions.....................................................................................................................30 Soil Characterization.......................................................................................................... ....31 Flushed Dariy Manure (FDM) Characterization....................................................................31 Analyses of P in Leachate.......................................................................................................31 5

PAGE 6

Soils and P Accumulation: FDM vs Phosphate......................................................................32 4 DISCUSSION................................................................................................................... ......42 Future Studies.........................................................................................................................48 Summary and Conclusions.....................................................................................................49 LITERATURE CITED..................................................................................................................49 BIOGRAPHICAL SKETCH.........................................................................................................54 6

PAGE 7

LIST OF TABLES Table page 3-1 Tavares soil description.....................................................................................................33 3-2 Millhopper soil description................................................................................................ 33 3-3 Orsino typical pedon description, taken from the Levy County Soil Survey Report........34 3-4 Characterization of so ils collected. (n = 3)........................................................................35 3-5 Sand size distribution among soils collected and pore volume of soil horizons used in the experiment................................................................................................................. ...35 3-6 Selected characteristics of raw and dige sted flushed dairy manure (FDM). (n = 3).........36 3-7 ANOVA output............................................................................................................... ...40 38 Duncans Multiple Range Test for Accumulated P...........................................................40 7

PAGE 8

LIST OF FIGURES Figure page 1-1 Schematic layout of University of Fl orida Dairy Research Unit manure-handling system; heavy arrows indicate flow of flushwater and manure solids...............................19 2-1 Columns containing 20g of soil were kept under 20 cm of suction utilizing Erlenmeyer flasks to ensure unsaturated flow of leachate, which was collected in glass vials.................................................................................................................... .......29 3-1 Correlation between total oxalate extracted Fe and Al in molar concentrations and RPA values of the soils......................................................................................................35 3-2 Background P of leachate obtained from add ition of 50mM KCl solution to soil columns .............................................................................................................................36 3-3 Total P in leachate for each pore volume...........................................................................37 3-4 Dissolved reactive P (DRP) concentrat ions in leachate for each pore volume..................38 3-5 Phosphorus in leachates for flushed da iry manure (FDM) treatments, shown as both total P and dissolved reactive P (DRP)..............................................................................39 3-6 Phosphorus accumulated in soils re ceiving different P amendments................................40 3-7 Correlation between P accumulated in soil columns, delivered as inorganic P at 48 mg L-1 P solution, and soil RPA values.............................................................................41 8

PAGE 9

Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science ENHANCED RETENTION OF PHOSPHORUS APPLIED AS FLUSHED DAIRY MANURE By Aaron Malek December 2007 Chair: Name Willie Harris Cochair: Ann Wilkie Major: Interdisciplinary Ecology Leaching of phosphorus (P) is a major concern in many areas of the world, especially in environments with karst topography that contain sensitive and valu able aquifers and springs. In one such area, North Florida of the SE United St ates, dairy farming is prevalent, and in such operations it is common to spray crop fields with flushed dair y manure (FDM) as a means of nutrient recycling. We investigated the fate of P in three sandy so ils that typify and encompass large areas of the region. At the University of Florida Dairy Res earch Center, anaerobic digestion of FDM is currently practiced and provides many benefits including reduc tion of organic matter (OM) and mineralization of nutrients, including P. Three different P amendments were applied to soil columns in a randomized complete block desi gn: raw FDM, digested FDM, and inorganic P. Since OM is known to form complexes and compet e for sorption sites with P, thereby reducing P sorption, it was hypothesized that re tardation of P would be least in soils receiving P as raw FDM and most in soils receiving inorganic P. Soil columns were kept under 20 cm of suction to promote unsaturated flow and amendments were applied at one half of the soil pore volume. Leachate was collected per pore volume and anal yzed colorimetrically for total P (TP) and dissolved reactive P (DRP). Littl e evidence of OM transport of P was found in columns receiving 9

PAGE 10

both raw and digested FDM, and through 30 pore volumes, very little leaching of P at all. The soils receiving inorganic P retarded leaching the least, and the extent of retardation was related to the soils inherent P sorbing capacity through Al and Fe oxide conten t. ANOVA output revealed significant differences in P accumulation in the so ils among P amendments and soils, as well as a P amendment soil interaction. D uncans Multiple Range Test showed the P accumulated in soils receiving inorganic P to be significantly less than those receiving raw or digested FDM. Precipitation with Ca and Mg contained in the FDM, entrapment of particulate P, and immobility of OM due to soil chemistry are the suspected explanations for the lack of P breakthrough in FDM amended columns. The precipitated P is not e xpected to be stable and will leach into the deeper soil horizons. This study highlights the diffe rences in P behavior in soils when delivered through different mediums and the complexity of the soil chemistry regarding P when FDM is amended to soils. It also demands further research in this area to better understand the processes and consequences of P leaching wh ere similar practices are employed. 10

PAGE 11

CHAPTER 1 INTRODUCTION Background Phosphorus (P) is an economica lly important input in both crop and livestock production systems (Hansen et al. 2002). It also poses an en vironmental risk if it migrates elsewhere in the environment. Increased concentrations of P in inland and coastal waters can lead to problems associated with eutrophication. Therefore, great importance has been given to determine the behavior and movement of P in the environmen t. Phosphorus is transported from agricultural fields via surface runoff and leaching. Leaching of P is especially important in sandy soils and even more important in soils overlying sensitiv e and valuable aquifers As part of nutrient management strategies, it is common practice on many dairy farms to spray fields with flushed manure in order to improve crop production and pasture land. However, many times high P levels occur in soils as a result of N:P imbalance in the wastewater. Studies by Iyamuremye et al (1996) and others highlig ht the importance of the interaction of P with organic matter. Dissolved organic matter may compete with P for sorption sites in soil, displace sorbed P, and mediate movement by sorption with P itself. Particulate organic matter may physically bloc k sorption sites, also promoting leaching of P. It follows that the application of inorganic phosphate may lead to less leaching of P than when applied in the presence of organics. Flushed dairy manure (F DM) wastewater has substantial amounts of suspended particulate and dissolved organic matt er. However, innovative techniques in treating FDM are increasingly used to improve farm operations in various ways. Fixed-filmed anaerobic digestion of FDM is one such technique, and it si gnificantly lowers and alters the organic matter content which may decrease leaching of P (Wilkie, 2004). Some of the benefits of anaerobic digesters include decreasing poten tial adverse effects the FDM may have on animal and plant life 11

PAGE 12

in associated water bodies by lowering biological oxygen demand (BOD), decreasing fecal coliforms, and production of methane. It is possibl e the digested effluent could also be used to minimize P leaching. This study seeks to determine if there is less retard ation of P when it is applied with organics compared to inorganic applicati on, and if so, whether there is a difference in retardation between anaerobical ly digested FDM and raw FDM. The results of this study will highlight differences, if any, between the moveme nt of inorganic and organic P, and contribute to the understanding of the role of organic matter in P mobility. Hypotheses The main hypothesis at the out set of this research was that organic matter in the FDM wastewater would compete with P for sorption site s in the soil, displace P on sorption sites, and mediate movement by sorption with P. Thus, soils applied with inorganic phosphate would retard P mobility the greatest. Because the digested FDM wastewater has less organic matter and organic P than the raw FDM wastewater, it should have greater retarda tion of P movement. The actual results of this st udy, however, documented behavior that directly contrasted with these initial expectations. Objectives Determine P leaching potential inorganic and total P when soil is amended with raw FDM wastewater, anaerobically digested FDM wastewater, and inorganic phosphate. Determine the relative P sorption capacity of sandy soil materials used in the experiment. Literature Review The movement of P through soils has been ex amined in many studies (Sharpley et al. 2003). The interactions and transformations of P within the soil profile are important in understanding the leaching risk of P. Studies ex amining different forms of P and P sources have also highlighted important differences in reactivity and leaching potential. Research has shown 12

PAGE 13

that P movement varies from soil to soil (Dodjic et al., 2004; Elrashidi et al., 2001; van Es et al., 2004) Many properties of soils have been s hown to have an impact on P behavior. The complexity resulting from the multiple factors de termining P behavior in soils has made it the subject of many studies. This review will exam ine literature concerning P movement when delivered in different forms, such as organic or inorganic fertilizers, the characteristics of different forms of P and its sources, especially ra w and digested FDM, and the interactions of P in the soil that relates to its retention and leaching potential. As the delivery and specific characteristics of P sources are important for understanding P fate, attention must be given to how it is derived, especially when considering organic amendments. The production of FDM is one opti on used by dairy operations to properly deal with livestock manure. Wilkie et al. (2004) descri be a system utilized at the Dairy Research Unit at the University of Florida in which the FDM is anaerobically digested in a fixed-film digester. Dairy facilities are kept sani tary by flushing the manure down na rrow alleys into a series of storage ponds where separation of bedding sand and very fibrous materials occurs mechanically and by sedimentation (Figure 1-1). The fixed-film di gester contains media on which microbes colonize. The microbes include a symbiotic group of anaerobic bacteria which convert complex organic matter into smaller molecules includ ing methane and carbon dioxide (Wilkie 2005). The benefits of anaerobic digestion include production of energy-yielding methane, mineralization of nutrients, including P, reduction of odors, inactivation of weed seed s, and lower pathogen levels. The reduction and change of the organic matter and mineralization of nutrients may have an effect on P mobilization when applied to the soil. Dissolved P interacts with Fe, Al, Mn, and Ca in soils (Brady and Weil, 2002; Pardo et al., 2003). Many authors have found the amount of Fe and Al in soils to be important in 13

PAGE 14

determining adsorption of P. Novak and Watts (2004) increased the sorp tion capacity of a sandy soil several fold by incorporating water treatment residuals that were flocculated with liquid alum (Al2(SO4)3). Freese et al. (1992) found phosphate adsorption in German soils to be predominantly related to the amount of amorphous Fe and Al oxides. Also, Phillips (2002) found leaching to be related to the availability of Fe oxides on soil colloids. Al-bound P was the main form of P when Elrashidi et al (2001) studied accumulation of P in Florida sandy soils (Myakka, Zolfo, and Adamsville). The P moved through the so il profile, but the lo w solubility of Alphosphates (Al-P) in the acidic saturated zone of these soil s prevented infiltration into groundwater. Ca-bound P (Ca-P) is also pH sensi tive but in some forms they are extremely insoluble and considered largely as unavaila ble (Brady and Weil, 2002; Pardo et al., 2003). Zhou et al. (1997) found the high adsorption capacity of the Spodosol Bh horizon a result of aluminum-organic matter complexes. Theses comp lexes can form in the soil profile and affect the P leaching potential as they release P mo re readily than inorganic metal oxides. Many studies have examined the interactions of P and other soil constituents, especially comparing these interactions when the P form is inorganic or orga nic. Most have found adsorption of inorganic P greater th an organic P, but not all. In fact, Leytem et al. (2002) found organic P forms to be preferentially adsorbed over inorganic, ortho-P. Th ey studied the behavior of four organic P compounds: three nucleot ides, ATP, ADP, AMP, and IHP (inositol hexaphosphate) while KH2PO4 was used as an inorganic reference. All the organic P compounds had greater adsorption than KH2PO4 on Blanton Sand and Cecil sandy clay loam. In the Belhaven sandy loam, IHP had the greatest adsorption followed by KH2PO4 and the nucleotides. When Lilienfein et al. (2004) studied preferenti al adsorption of organic and inorganic P, the organic P already present in the soil was exam ined, not specific organic P compounds. They 14

PAGE 15

found preferential adsorption of PO4 over dissolved organic P. B ecause the number of phosphate groups on a molecule increases its sorption, and the number and density of phosphate groups in soil organic matter is variable, the adsorption ch aracteristics vary as well, which explains differing results when the P source and form ch anges (Qualls and Haines, 1991; Leytem et al., 2002). Leaching of P when using organic and inorga nic fertilizers is ofte n studied. Siddique et al. (2000) conducted leaching trials using five loam soils with P from anaerobically digested sewage, processed into dry biosol ids, or inorganic Ca-P. Both s ources of P were amended into the different soils and placed in soil columns (6.5 cm diameter and length of 30 cm) to a depth of 20cm. Although the biosolids contain organic P, af ter the digestion process, up to 80% of P is inorganic, and up to 97% of the P leached was i norganic. The columns were leached with a total of 5 L of deionized water. In most cases more P was released from the inorganic fertilized soils. This was attributed to slower P saturation in th e biosolid treated soils caused by less soluble P in the biosolid amendments compared with the Ca-P fertilizer. Jiao et al. (2004) compared loads of P from soils receiving either organic or inorganic fertilizer. The soil was a fine-silty, mixed, frigid Typic Endoaquent. The composition of the soil was 300 g kg sand, 540 g kg silt, and 160 g kg clay with 15.4 g total C kg and pH 6.1 in the 0to 15-cm layer. Soil columns (10 cm diameter and 20 cm depth) were leached with synthetic rainwater and nutrient loads were calculated. Soils receiving inorganic fertilizer ha d 48% less dissolved reactive P load than soils receiving organic fertilizer. Th ey stated that the soil had le ss sorption capacity for dissolved organic P than for inorganic P. The dissolved reactive load was positively related to the soil Mehlich-3 P concentration (R2 = 0.50). Carefoot and Whalen (2003) studied leaching of P from a silt-loam Gleysol fertilized with inorganic (tripl e superphosphate) and orga nic (composted cattle 15

PAGE 16

manure) P sources by measuring P in the subsurface water at a 60cm depth. It contained 0.3 to 1.7 mg P L-1. Particulate P was the dominant P form at most sampling dates. Also, Phillips (2002) studied leaching of P from piggery wast ewater using a Vertisol and a Spodosol in undisturbed soil cores (diameter of 30 cm and de pth of 60 cm). Both molybdate reactive P and unreactive P were leached. The author lists unreactiv e P as dissolved organic P, particulate P, and non-reactive P. The Spodosol leached mos tly molybdate reactive P (approximately 70%) because the wastewater containe d about 70% of this form, and the soil had little adsorption capacity. The P leached from the Vertisol was mostly (approximately 80%) unreactive P because the molybdate reactive P was adsorbed by the soil colloids. These studies highlight differences in leaching potential of P from soils when different forms of inorganic and organic P are applied to different soils. The interactions of organic matter with P and P adsorption sites are a critical area of research concerning P mobility. Organic co lloids have been shown to transport metal contaminants in subsurface soils. Karathan asis and Johnson (2006) reported higher metal elutions up to four orders of magnitudes greate r than the controls. Biosolids, derived from municipal wastes, and poultry manure were applied to an Alfisol, a Mollisol, and an Entisol. The metals, Cd, Mo, and Cr, were present in both the particulate and sol uble fractions, and the significant increase was attributed to increa sed organic-metal complexation and exclusion facilitators. Iyamuremye et al. (1996) found that organic am endments to soil decreased P sorption sites and sorption that was related to changes in pH an d exchangeable Al. Chardon et al. (1997) conducted a set of experiments examining the leaching potential of dissolved organic P (DOP) in cattle slurry applied on sandy soil. Th e DOP in the slurry consisted mainly of high molecular weight compounds. Leaching of P was examined using laboratory soil columns 16

PAGE 17

(diameter of 15.3 cm and length of 100 cm) and outdoor lysimeters. The soil columns contained a quartzitic sand, with a total P of 6.5 mg kg-1, up to 70 cm with a 16 cm layer of a sandy loam, containing 530 mg kg-1, total P placed on top. The columns we re leached with 100 mL of water twice a week for four months. Of the total P le ached, more than 90% from the soil columns and more than 70% from the lysimeters leached as DOP, indicating its high mobility. Also, DOP in the soil pore water increased from about 10% of total P in the topsoil to more than 70% at 70 to 80 cm depth. Because the leaching of total P mainly occurred in periods of low Cl and NO3 concentrations, indicative of high leaching rates, the authors suggested that the P transport was mediated by dissolved organic carbon and other co lloidal particles. Eghball et al. (1996) also found deeper movement of P in soils receiving manure application than soils receiving equal amounts of inorganic P. They, t oo, suggested that the P moved in organic forms or chemically reacted with compounds in manure, enhancing solubility. This was examined further by Motoshita et al. (2003). They studied leaching of colloidal and dissolved P in soil column experiments using a surface loam soil with a hi gh Olsen-P content (93 mg-P/kg). The columns were leached with artificial ir rigation solution. Colloidal P leaching showed a minor increase with time, and dissolved P leaching was nearly constant. Dissolved P consisted of 81-86% of total P leached. There was a high correlation betw een dissolved P leaching and dissolved organic matter leaching (R2 = 0.82). The P sorption was investigated and showed that the P was sorbed to or formed complexes with the dissolved organic matter. These studies show the importance of organic forms of P and dissolved organic matter relating to P mobility in the soil. In addition to the interactions affecting P mobility, many have investigated the physical pathway and flow that P takes within the soil. This is especially important for finer textured soils. Djodjic et al. (2004) sought to establish a relationship be tween soil P levels and actual P 17

PAGE 18

leaching in structured clay soils. Leaching of total P and dissolved reactive P were measured over three years in undisturbed soil columns. There was no general correlation between P concentrations and soil test P or P sorption indices of the topsoil. In one soil, where preferential flow was the dominant water transport pathwa y, water and P bypassed th e high sorption capacity of the soil, resulting in high P losses. A compar ison of two soil textural extremes, a clay loam and loamy sand, was conducted by van Es et al. (2004). The study took place on farm land with artificial drainage a nd liquid manure application. High P l eaching losses were measured in the clay loam as soon as drain lines initiated flow after manure app lication. Flow weighted mean P leaching losses on clay loam plots averaged 39 times higher than those on loamy sand plots. Preferential flow was determined to be the main transport mechanism in the clay loam, but the authors stated P leaching from manure applications on loamy sand soils do not pose environmental concerns as long as soil P levels remain below saturation levels. This is further supported by Akhtar et al. (2003) where five so ils of differing textures were leached with synthetic acid rainwater enriched with inorganic and organic P. At low flow rates, P appeared in the drainage water soon after applic ation of either inorganic or organic P for the silt loam soil. Soils in which matrix type flow dominated had li ttle or no increase in drai nage water P. Elevated P concentrations in the drainage water could not be explained by the P adsorption strength of the soils with the possible exception of the sandy loam soil, where the outflow P concentration was consistently low. Agricultural practices, such as plowing, affect the porosity of the soil. Geohring et al. (2001) found that plowing in liquid manure appli cations before irrigation greatly reduced P leaching. Column studies using pack soil and arti ficial macropores were designed to examine the role of macropore size on P sorption to pore walls. They found that soluble P may be transported 18

PAGE 19

through macropores 1mm or greater with negligib le P sorption to pore walls. When macropores were absent, no measurable P was transporte d through the soil colum n. The macropores were disrupted from plowing which promoted matrix flow and reduced P leaching. The literature, to date, explains many of the factors concerning the potential for P mobility in soils. With regards to sandy soils, stud ies have shown variable ability to uptake P. Some of the sandy soils had littl e capacity while others almost completely prevented leaching. This is due, in part, to the inherent P sorbing compounds present in the soil, such as Fe and Al oxides. When organic amendments are added to the soil, complexes that mediate transfer of P can form. The extent that this occurs varies among soils and also depends on the P source. Organic sources of P, such as raw and digested FDM, can have different effects on P mobility than other organic sources. The consequences of excessive P leaching, such as eutrophication, require an understanding of all the contributing factors. Furthe r research of P movement in various soils and application methods will aid in the development of management practices that both aid in agricultural producti on and prevent wa ter pollution. Figure 1-1. Schematic layout of University of Florida Dairy Research Unit manure-handling system. Heavy arrows indicate flow of flushwater and manure solids. A) Milking parlor. B) Freestall barns. C) Flushwater holding tanks. D) Wa stewater collection channel. E) Sand-trap. F) Mechanical sepa rator. G) Sedimenta tion basin. H) sampling Pit. I) Primary storage pond. (Wilkie, 2004) 19

PAGE 20

CHAPTER 2 MATERIALS AND METHODS Soil Selection and Sampling Sandy horizons were sampled from soils of three series that o ccur in the coastal plain of the southeastern USA: Tavares (sandy, hyperthermic, Typic Quartzipsamment), Millhopper (loamy, siliceous, hyperthermic, Grossareni c Paleudult), and Orsino (sandy, hyperthermic, Spodic Quartzipsamment). Tavares and Millhopper soils are extensive in Florida across a range of moderately-well-drained landscapes, as are th e taxonomic families they represent. Orsino soils are commonly associated with old dunes or other landforms underlain by deep sandy parent materials and occupied by oak scrub or sa nd pine plant communities. These three soils encompass a wide range of sandy soil horizons and environments in Florida, and were selected on this basis. Other research crite ria satisfied by these soils is that they (i) prospectively provide an intermediate and representativ e range of P retardation and ( ii) they occur on some leachingprone landscapes which are the types of soil/landscapes to which results of this study should apply. The E2 horizons of the Millhopper and Ta vares and the Bw horizon of the Orsino were chosen for the column experiment described belo w. All soils were air dried and sieved through a 2 mm sieve prior to experimental use. Soil Analyses Oxalate-Extractable Iron (F e) and Aluminum (Al) The Fe and Al content of the soils was determined by extraction with acid ammonium oxalate in the dark (Baril and Bitton, 1967; McKe ague and Day 1966). This method extracts both amorphous inorganic Fe and Al a nd organic complexed Fe and Al in soils. The acid oxalate extracting solution was made by mixing 700 mL of 0.2 M ammonium oxalate and 535 mL of 0.2 M oxalic acid together and adjusting pH to 3.0 by using either of the base solutions. For each 20

PAGE 21

sample, 2 g of soil was placed in 50 mL centrifuge tubes that were wrapped in aluminum foil to exclude light. A 50-mL aliquot of acid oxalate extraction solution was added to the soil in tubes, which were then capped tightly and placed in a re ciprocal shaker at low speed for 4 hours. Tubes were centrifuged for 20 min at 2000 rpm, and su pernatant solutions filtered through Whatman 42 filters and stored cold (4C) unt il analysis within a few days. An atomic absorption spectrophotometer was us ed to determine Fe and Al content. The standard solutions were prepared so that the matrix contained the same concentration of acid oxalate as the extracting solutions. Total P (TP) of Soils The TP of the three soils was determined by using an ashing and acid digestion procedure (Anderson, 1976). A 50 mL beaker containing 1 g of soil was placed into a muffle furnace at 350C for one hour and at 550C for two more hours. After cooling, 20 mL of 6M HCl was added, and it was allowed to slowly evaporat e on a hotplate. After the sample dried, the temperature was raised briefly and then allowed to cool. Then 2.2 mL of 6M HCl was added, and the beakers were heated so that the residue was easily dislodged. The mixture was then transferred to a funnel with a Whatman #42 filter paper, and the solution filtered into 50 mL volumetric flask. The beakers and the filters were ri nsed repeatedly with deionized water, and the volume of the flasks was brought up to 50 mL. Analysis of P in extracts was done colori metrically (Murphy and Riley, 1962), with P standards ranging from 0 to 35 mg L-1. 0.5 mL of each standard and unknown solution was placed in a test tube. Reagent As was prepared by dissolving 12 g of ammonium molybdate in 100 mL of H2O and 0.2908 g of antimony potassium ta rtrate in another 100 mL of H2O; both of these solutions were added to 1 L of 2.5M H2SO4 and diluted to 2 L. This was kept stored in a 21

PAGE 22

dark, cool place. Reagent Bs was made fresh from a st ock solution of Reagent As for each analysis. To make Reagent Bs, 1.056 g of ascorbic acid was disso lved in 200 mL of Reagent As. Reagent Bs was added in 4 mL aliquots to test tubes containing soil extractant solutions or standards. Then, 10 mL of deionized H2O was added while turning the tube to rinse the sides. Blue color developed in the test tubes commensu rate with the amount of P they contained. The absorbance readings were taken on a spectr ophotometer at 880 nm wavelength after color developed for thirty minutes. The st andard curve was accepted if the R2 was at least 0.997. The absorbance readings of the soil samples were in terpolated on the standard curve to determine P mg L-1. Relative Phosphorus Retention Capacity (RPA) The RPA is a procedure developed to provide a quick practical measure of the P sorption capacity of sandy soils (Rhue et al, 1994). This analysis used Reagent A and Reagent B that differs somewhat from the Reagent A and Reagen t B that was used in determining TP of the soils. For convenience, the reagents used in the TP of soil procedure will be referred to as Reagent As and Reagent Bs, and the reagents used in this and other analyses will be referred to as Reagent Aw and Reagent Bw. Reagent Aw was prepared by combining three different solutions: 12 g of ammonium molybdate dissolved in 200 mL of distilled water, 0.2908 g of antimony potassium tartrate dissolved in 50 mL of distille d water, and 144 mL concentrated sulfuric acid (18M) in 1500 mL of distilled water that was allowed to cool. The solution mixture was then diluted to 2 L, mixed, and stored in a cool, dark place. Prior to each analysis, Reagent Bw was made by adding 0.75 g of ascorbic acid to 50 mL of Reagent Aw and diluting to 500 mL. The RPA was determined by placing 10 g of soil in 20 mL scintillation vials and adding 2 mL of 2000 mg L-1 P solution. The vials were capped and vigorously shaken. This was allowed to equilibrate for 24 hours. Whatman #1 filters were cut to fill the bottom of 50 mL polyethylene 22

PAGE 23

centrifuge tubes that had 6 to 8 small holes in the bottom. The filter was also wetted with 50 mM KCl and centrifuged by itself at 1500 g for 5 mi n to drain excess KCl solution. The wet soil was transferred to the centrifuge tube s and collection cups were taped to the bottom. The tubes were then centrifuged at 1500 g for 5 min in order to extract the soil soluti on. The solution was then filtered through 0.45 m filters, and 0.1 mL aliquots were put in glass centrifuge tubes. A set of P standards were made by transferri ng 0.1 mL of 0, 50, 100, 200, 400, 600, 800, 1000, 1500, and 2000 mg L-1 P standards in glass centrifuge tubes. All were placed in a drying oven at 70C overnight. After drying, 20 mL of 0.1M HNO3 was added to all tubes. The tubes were mixed vigorously and 0.5 mL aliquots we re added to 5 mL Reagent Bw. Color development proceeded for two hours, at which time absorban ce readings were taken at 880 nm. The amount of P adsorbed (S) was ca lculated using the formula: S = [(2000 Cps) x 2] / 10, where S is the amount of P adsorbed ( g/g of soil), Cps is the P concentration in pore solution after 24 hour equilibra tion with soil ( g/mL). With S, the RPA was calculated thusly: RPA = S/400, where 400 g/g represents the maximum adsorption possi ble under the experiment al conditions. Particle Size Distribution Analysis (PSDA) of Soils The particle size distribution for each soil wa s determined (Soil Survey Staff, 1992). To begin, 100 mL of Calgon (5% so lution of sodium hexametaphosphate) was added to 250 mL Erlenmeyer flasks containing 50 g of dried so il. The flasks were stoppered and placed on a shaker for 16 hours. The contents were transfer red into cylinders and were diluted to 1000 mL with deionized water while thorough ly rinsing the flasks. The cylinders were shaken and inverted 6 times. They were then placed in a bath at 3 min intervals. After the allo tted time, a glass pipette was lowered into the cylinder at the allotted depth to withdraw the clay fraction. The clay 23

PAGE 24

aliquots were placed in weighed, metal cups, dr ied in an oven, and weighed again the following day. The contents of the cylinders were then rinsed through a No. 325 sieve (48-m mesh) to remove the clay and silt. The remaining sand wa s collected and after dr ying, it was fractionated into very coarse (2000 -1000 m), coarse (100-500 m), medium (500-250 m), fine (250-100 m), and very fine (100-48 m particles using sieve Nos. 18, 35, 60, 140, and 325. Sand fractions were weighed and the distribu tion was calculated using the formulas: %Sand = (sand weight) (100) / (sample weight), %Clay = {[(metal cup + clay weight + Calgon) metal cup blank weight](40)}(100) / (sample weight), %Silt = 100% %Clay %Sand. Flushed Dairy Manure Analyses Flushed dairy manure samples were collected at the Dairy Research Unit in Hague, Florida. The raw FDM was collected from the pu mp sump that feeds into the digester. The digested FDM was obtained directly from the effl uent line. All samples were stored in a cold room at 4C. The FDM samples were primarily analyzed for TP, DRP, pH, and chemical oxygen demand (COD). The samples from September 14, 2005 were chosen as treatments for this study out of all samples taken because the digested and raw FDM show ed typical differences in DRP and COD, i.e., DRP was greater, and COD less, for digested FDM. It was important for the purposes of this study that digested and ra w FDM show these differences because the hypothesized differences in potential P leaching re lates to the proportion of DRP to TP. The pH was measured with a standard, portable pH me ter. The TP was determined by taking a 5 mL aliquot of wastewater and diluting to 10 mL with deionized wate r. From this mixture, 0.1 mL was taken and placed in a Hach COD reactor test tube. This was done so no more than 2.5 g of P would be contained in each test tube. The samples were dried and then ashed in a muffle furnace at 300C for 1 hour and at 550C for tw o more hours. Aliquots of 0.1 mL P standard 24

PAGE 25

solutions of 0, 5, 10, 15, 20, 25, and 30 mg L-1 P were transferred into test tubes and were evaporated. Then, 5 mL of Reagent Bw was added to each sample and standard, and blue color developed for two hours at which time abso rbance readings were taken at 880 nm. Total P (TP) Method Investigation HCl Digestion vs Only Ashing Five mL of FDM was diluted to 10 mL w ith deionized water. After mixing, 0.1 mL aliquots, totaling 12, were transferred to Hach COD reactor test tubes. All test tubes were dried using the COD reactor at 120C. After drying, all te st tubes were placed in a muffle furnace at 300C for 1 hour and 550C for two hours. After cooling, 6 of the wastewater samples, two of the 30 mg L-1 P standards, and one of the 0 mg L-1 P standard received 5 mL 6 M HCl. These were heated in the COD reactor for several hours at 120C, just above refl uxing temperature but, no visible signs of boiling, until all liquid had ev aporated. After cooling, 5 mL of Reagent Bw was added to each sample and a set of standards, and blue color developed for 2 hours at which time absorbance readings were taken at 880 nm. The P concentrations of the samples that were HCl digested with 6 M HCl were compared with thos e that were not. It was determined that this modified procedure was satisfact ory for the purposes of this st udy, enabling rapid assessment of TP in small volumes of column effluent (descr ibed below). The mean TP of the FDM that had HCl digestion was 46.5 0.6 mg L-1 and the mean TP of the FDM with no HCl digestion was 48.6 0.8 mg L-1. Dissolved Reactive Phosphorus (DRP) In order to determine if any constituent of the wastewater interfered with analysis reagents and skewed results, a spike recovery test was performed. The raw and digested FDM was filtered through 0.45 m filter, and 0.5 mL was diluted to 15 mL with distilled water. From this, 1 mL aliquots were transferred into test t ubes. In duplicate, samples were spiked with 0.1 mL of 0, 5, 10, and 15 mg L-1 P standards. A set of standards was made: 0.1 mL of 0, 5, 10, 15, 25

PAGE 26

20, 25, and 30 mg L-1 P. Five mL of Reagent Bw was added to each sample and standard, and blue color developed for two hours at which time absorbance readings were taken at 880 nm. The increasing spike was checked to be linear (r2 > 0.998). Chemical Oxygen Demand (COD) The chemical oxygen demand (COD) of the raw and digested FDM was determined as an indication of digestion. The mg L-1 COD results are defined as the mg of O2 consumed per liter of sample used in the procedure. The anaerobic digestion process eliminates a significant amount of the COD from the raw FDM. COD analyses we re performed as described in Wilkie (2004). First, 500 mL of each FDM was homogenized in a blender for 2 minutes, and a 2 mL aliquot was added to the COD Digestion R eagent Vial provided by Hach. Also, a subsample of the homogenized FDM was centrifuged at 12,000 min-1 for 30 min, and the supernantant was sampled to determine the soluble COD. Vials we re heated in the Hach COD reactor for 2 hours at 150C. The reagent contains dichromate (Cr2O7 -2), which oxidizes organic compounds and is reduced to the green chromic ion (Cr+3). The amount of Cr+3 produced is determined colorimetrically from which the COD is calculated. Soil Column Experimental Procedures Column Construction Twenty seven cylindrical soil columns were used in this experiment. They have an inside diameter of 1.6 cm and stand 20 cm tall. The colu mns have a bushing-like fitting that contains a glass frit at the bottom. A nippl e at the bottom and a piece of glass tubing are made continuous by a piece of clear plastic tubing. The collecti on vial encompasses the glass tubing so that suction can be applied to the vial, pulling the soil leachate under unsaturated flow. The continuous suction is supplied by an elevated Erlenmeyer flask th at has a siphon held to it. Air leaks anywhere on the column are mediated by tape and petroleum sealant. The elevation of the 26

PAGE 27

Erlenmeyer flask controls the amount of suction. Fo r this experiment the flasks were held at an elevation that results in 20 cm of suction (Figure 2-1). Pore Volume Determination The mass of soil used in each column for this experiment was 20 g. The pore volume for each soil was determined by placing 20 g of soil in weighed columns, wetting with 50 mM KCl, and applying 20 cm of suction. When excess KCl finished passing through the column and any addition of KCl resulted in im mediate elution, the column was weighed again to determine the pore volume. Background P The background P of the soils was determin ed by elution with 50 mM KCl. The first background value was collected from leachate fr om the initial wetting of the soil. The subsequent background values were taken from leachate obtained by the addition of 0.5 pore volumes of KCl to the columns at least one da y apart. A total of 6 KCl pore volumes were collected to get steady background P values. The n, 2.5 mL aliquots of the soil leachates were filtered through 0.45 m filter, placed in test tubes, and dried in an oven overnight at 70 C. A set of standards: 0.1 mL of 0, 1, 2, 3, 4, and 5 mg L-1 P was made. Each sample and standard received 5 mL of Reagent Bw, and color developed for two hours at which time absorbance readings were taken at 880 nm. Flushed Dairy Manure (FDM) Storage and Sampling Approximately 11.5 liters of both raw FDM a nd digested FDM were collected for the experiment. They were collected and stored in 3.8 L jugs at 4C. A procedure was devised to minimize anticipated clogging of soil columns a nd to provide a regular consistency of the FDM amendments when applied to soil columns a nd when sampled for analytical purposes. The largest suspended particles were allowed to settle at the bottom of the jugs. The smaller particles 27

PAGE 28

remained suspended. At the time of each addition, the jugs were slightly shaken, in a similar way, to mix the suspended particles and then a sm all portion was transferred to a separate beaker. Each jug was not used once it had become halfwa y empty. This ensured an approximately equal consistency of the FDM added to the soil columns that was evident by observation. Leaching Procedure Each soil received three trea tments: raw FDM, digested FDM, and a solution of KH2PO4. Based on TP analyses the solution of KH2PO4 was prepared at 48 mg L-1 P with a 50 mmol KCl background electrolyte concentration. This ensure d roughly equal TP values for all amendments. Treatments were done in triplicate per soil in a randomized complete block design. To allow for adequate sorption of P, the rate of addition was 0.5 pore volume of the amendment per addition. The leachate was collected per whole pore volume. The rate of the additions was limited by clogging of some columns receiving FDM amendments. In some cases more than 48 hours elapsed for complete infiltration. The clogging occurred in the top portion of th e columns as the larger particulates settled there and slowed infiltration. The columns were kept in dark as mu ch as possible to discourage algal growth. Leachate Analyses The soil leachates were analyzed for TP a nd DRP. Total P determination was performed by the ashing procedure without HCl digest ion, as described in the Total P method investigation section. Dissolved reactive P anal ysis included filtration, dilution if needed, and the addition of Reagent Bw. Absorbance was taken at 880nm after 2 hours. Statistical Analyses An ANOVA was run evaluating the statistical significance of the randomized complete block design experiment, and Duncans Multiple Range Test was used to determine significant differences in the P accumulated in the columns among the three treatments. Both of the 28

PAGE 29

preceding tests utilized SAS software. A regression analysis was used to demonstrate the significance of the relationship of P accumulated in the columns receiving P as PO4 and the RPA value of the soil. Figure 2-1. Columns containing 20g of soil were kept under 20 cm of suc tion utilizing Erlenmeyer flasks to ensure unsaturated flow of leachate, which was collected in glass vials. 29

PAGE 30

CHAPTER 3 RESULTS Soil Descriptions The Tavares sand was collected on May 18, 2005 in Alachua County, FL 604m south of Lake Mize in the Austin Cary Memorial Forest The soil was collected by boring three holes, 20 cm apart, with an auger. Tavares sand is deri ved of sandy marine deposits and is moderately well drained. It occurs in a forested environment on a <2% slope dominated by oak and long-leaf pine. The Tavares soil horizons are described in Table 3-1. The Millhopper soil was collected on May 18, 2005 in Alachua County, FL. University of Florida, 126 m WNW (307) fr om the NW corner of Museum Rd and Newell Dr. The soil was similarly collected by boring three holes, 20 cm apart, with an a uger. It is a moderately well drained soil derived from sandy ma rine deposits. It occurs in a small forested area on a <5% slope dominated by sweetgum, hickory, oak, and ot her small trees and shrubs. Table 3-2 shows the descriptions of the Millhopper soil horizons. The Orsino soil Bw horizon was obtained from the Soil Mineralogy lab at the University of Florida. It was collected by Willie Harris from a vertical exposure of an old sand dune near Cedar Key (Levy County), FL, during a professional field trip. It occurred in an Orsino fine sand map unit and the profile conformed to the Orsino series (personal communication, Willie Harris). Orsino soils typically occur on elevated ridge s and knolls. They have low available water capacity and fertility, but still support natural vegetation includ ing slash pine, sand pine and scrub oak. A description of this soil taken fr om the Levy County Soil Survey (Slabaugh et al., 1996) is included in Table 3-3. The typical pedon is listed as be ing located 3,300 feet (approximately 1000 m) north and 250 feet east (approximately 76 m) of the southwest corner of sec. 11, T. 14 S., R. 16E. The Bw horizon used in this study was morphologically similar to the 30

PAGE 31

Bw1 of the Levy County typical pedon, with the exception that th e matrix color was 10YR 5/6 instead of 10YR 6/4. Soil Characterization The results of the analyses characterizing the soils in Table 3-4 show some important properties for P adsorption and mobility in sandy so ils. The TP of the three soils were typical and indicated little or no impact fr om anthropogenic activities regard ing P loads. The RPA varied considerably among the soils with the Tavares soil having an une xpectedly high value, but there was a commensurate increase in the amount of Fe and Al oxide present, and there was high correlation (r2 = 0.90) between the RPA and the total amount of oxalate extracted Fe and Al (Figure3-1) The Orsino Bw horizon had the high est percentage of fine sand (Table 3-5), consistent with its formation in dunal parent mate rial which characteristically is well sorted and dominated by fine sand. Note that the Tavares E1 and Millhopper E1 were not used in the column experiment. Flushed Dairy Manure (FDM) Characterization Mean TP values of the raw and digested FDM are similar whereas the mean amount of DRP in the digested FDM is about 25% more than in the raw FDM (Table 3-6). The spike recovery test results showed no interference in the DRP testing procedure. The digestion process removed over half the COD present in the wastewat er, but did not appear to significantly change the pH. Analyses of P in Leachate Background P values (Figure 3-2) stabilized after 6 pore volumes. Less than 0.1 mg L-1 P was detected in any horizon with the Millhopper E2 horizon having slightly higher values. Soil differed with respect to retardation of P applie d as inorganic P (Figure 3-3). The Millhopper soil started to have breakthroughs of P on the second pore volume. The Orsino and Tavares soils 31

PAGE 32

adsorbed virtually all P until the 9th and 19th pore volume, respectively. This order of P adsorption capacity is in agreement with the RPA values of the soil horizons. In all soils, there were higher values of TP in the leachate for the first few pore volumes. These values decreased until the 5th pore volume. The largest difference of TP between raw and digested FDM amended soils was in the Millhopper soil where le achate from the digested FDM was slightly higher than those derived from the raw FDM. The leachate of the Tavares soil derived from the raw and digested FDM showed little differences in TP values. There were also little differences from the Orsino soil, except for the leachate from the raw FDM having higher values initially and the digested FDM leachate being higher at the la st pore volumes. In all cases, after the breakthrough of P derived from the inor ganic P treatment occurred, the level of TP in the leachate derived from the raw and digest ed FDM was lower than those derived from inorganic P. The DRP in the leachate roughly follows the same trends as the TP (Figures 3-3, 3-4, and 3-5). Leachates derived from the raw and digest ed FDM had low levels of DRP for the Tavares and Orsino soils, except for an increase in the la st pore volumes derived from the digested FDM in the Orsino soil. The DRP in the leachate from the raw and digested FDM from the Millhopper soil is higher than other soils. Ther e is an increase in the last por e volumes as well as a spike in concentration at the 20th pore volume. The largest differences between TP and DRP values occur during the initial por e volumes (Figure 3-5). Afterwards the levels of P are more closely aligned, especially for Tavares and Orsino. Soils and P Accumulation: FDM vs Phosphate Calculated P accumulation in the soils (Figur e 3-6) shows that the raw and digested FDM accumulate similarly in each soil and among all soils and to a significantly greater extent than for the inorganic P treatment (Tables 3-7 and 3-8) These values were derived by subtraction of P 32

PAGE 33

in leachate from that which was added to the columns. The P delivered as inorganic P accumulated differently and relates to the RP A (Figure 3-7), which was found to be well correlated with the amount of oxalate extractable Fe and Al oxides (Figure 3-1). Table 3-7 shows the accumulation of P in the soil columns only differ significantly be tween the inorganic P amendments and the FDM amendments. Differences between soils were not tested statistically due to the soil x amendment inte raction (Tables 3-7 and 3-8). Table 3-1. Tavares soil description Horizon Depth (cm) Description A 0-15 Dark grayish brown (10YR 4/2) sand, clear boundary. AE 15-50 Brown (10YR 5/3) sand, gradual boundary. E1 50-75 Brown (10YR 5/3) sand, gradual boundary. E2 75-95 Light yellowish brown (10YR 6/4) sand, gradual boundary. E3 95-120 Pale brown (10YR 6/3) sand, redox concentrations and depletions, gradual boundary. E4 120-200 Very pale brown (10YR 7/3) sand, redox concentrations and depletions, gradual boundary. Table 3-2. Millhopper soil description Horizon Depth (cm) Description A 0-14 Very dark grayis h brown (10YR 3/2) sand, clear boundary. AE 14-29 Brown (10YR 4/3) sand, clear boundary. E1 29-52 Yellowish brown (10YR 5/4) sand, gradual boundary. E2 52-83 Brown (10YR 5/3) sand, gradual boundary. E3 83-105 Pale brown (10YR 6/3) sand, gradual boundary. E4 105-118 Light gray (10YR 7/2) sand, redox concentrations and depletions, red matrix, clear boundary. Bt 118-140 Brown (7.5YR 5/4) sandy loam, redox concentrations and depletions, clear boundary. Btg1 140-175 Light gray (10YR 7/1) sandy cl ay loam, redox concentrations and depletions, gradual boundary. Btg2 175-189 Light gray (10YR 7/1) sandy clay, redox concentrations and depletions. 33

PAGE 34

Table 3-3. Orsino typical pedon description, taken from the Levy County Soil Survey Report (Slabaugh et al., 1996). The Orsino Bw horizon used in this study was collected from a similar Orsino soil in Levy County. Horizon Depth (cm) Description A 0-10 Gray (10YR 5/1) fine sand; weak fine granular structure; very friable; many fine roots, st rongly acid; clear wavy boundary. E1 10-20 Very pale brown (10YR 7/ 2) fine sand; singl e grained; loose; strongly acid; clear wavy boundary. E2 20-32 White (10YR 8/1) fine sand; single grained; loose; strongly acid; abrupt irregular boundary. Bw and Bh 32-122 Brownish yellow (10YR 6/6) fi ne sand (Bw); single grained; loose; common fine roots; discontinuous lenses of weaklycemented dark yellowish brown ( 10YR 4/4) fine sand (Bh) that are 1 to 5 cm thick at the upper co ntact of the horizon; strongly acid; gradual wavy boundary. Bw1 122-147 Light yellowish brown ( 10YR 6/4) fine sand; few fine faint brownish yellow (10YR 6/6) mottles; single grained; loose; few fine roots; strongly ac id, gradual wavy boundary. Bw2 147-175 Brownish yellow (10YR 6/8) fine sand; single grained; common fine distinct strong brown (7.5YR 5/8) mottles; loose; few fine roots; strongly acid; gradual wavy boundary. C 175-200 White (10YR 8/1) fine sa nd; few medium distinct yellow (10YR 7/8 mottles; single grained; loose; moderately acid. 34

PAGE 35

Figure 3-1. Correlation between to tal oxalate extracted Fe and Al in molar concentrations and RPA values of the soils. Data points incl ude soils Millhopper E2, Orsino, Millhopper E1, Tavares E2, and Tavares E1 in order of ascending RPA (Table 3-4). Table 3-4. Characterization of soils collected. (n = 3) Soil TP mg kg-1 RPA (%) Fe mg kg-1 Al mg kg-1 Total Fe & Al mmol kg-1 Sand g kg-1 Silt g kg-1 Clay g kg-1 pH Orsino Bw 177 67 544 857 41.5 980 10 10 4.0 Tavares E1 208 97 811 1380 65.7 950 5.8 Tavares E2 200 91 856 1008 52.7 950 30 20 5.7 Millhopper E1 231 78 279 919 39.0 940 40 20 6.3 Millhopper E2 169 49 324 511 24.7 950 30 20 6.0 oxalate extracted clay was lost, so only sand is reported Table 3-5. Sand size distribution among soils collected and pore volume of soil horizons used in the experiment. Soil % Very Coarse % Coarse % Medium % Fine % Very Fine Pore volume at 20 cm suction (mL) Orsino .2 3.0 22.6 66.7 5.3 5.0 Tavares E1 .3 7.0 44.5 36.2 6.9 Tavares E2 .3 6.1 42.4 37.7 8.5 4.2 Millhopper E1 .3 5.8 43.9 40.3 3.5 Millhopper E2 .3 5.7 45.9 39.3 4.1 4.7 35

PAGE 36

Table 3-6. Selected characteristics of raw and digested flushed dairy manure (FDM). (n = 3) Total P mg L-1 Dissolved Reactive P mg L-1 Dissolved Organic P mg L-1 % DRP Total COD mg L-1 Soluble COD mg L-1 pH Raw FDM 45.8 29.5 1.0 64 3360 1770 7.4 Digested FDM 48.9 39.1 1.0 80 1595 696 7.3 COD = Chemical oxygen demand Figure 3-2. Background P of leachate obtained fr om addition of 50mM KCl solution to soil columns 36

PAGE 37

Figure 3-3. Total P (TP) in leachate for each por e volume. Definite breakthroughs can be seen in the columns receiving the inorganic P amen dments contrasted with the lack of breakthroughs in the columns receiving the flushed dairy manure (FDM) amendments. A) Millhopper TP. B) Tavares TP. C) Orsino TP. 37

PAGE 38

Figure 3-4. Dissolved reactive P (DRP) concentr ations in leachate for each pore volume. With regards to flushed dairy manure (FDM) am ended columns, concentrations DRP and total P are similar which suggests that near ly all total P leaching through the columns is DRP. A) Millhopper DRP. B) Tavares DRP. C) Orsino DRP. 38

PAGE 39

Figure 3-5. P in leachates for flushed dairy manure (FDM) treatments, shown as both total P (TP) and dissolved reactive P (DRP). With the exception of the beginning pore volumes, this illustrates that P in the l eachate is largely comprised of DRP. A) Millhopper raw FDM: TP vs. DRP. B) Millh opper digested FDM: TP vs. DRP. C) Orsino raw FDM: TP vs. DRP. D) Orsino di gested FDM: TP vs. DRP. E) Tavares raw FDM: TP vs. DRP. F) Tavare s digested FDM: TP vs. DRP. 39

PAGE 40

Figure 3-6. P accumulated in soils receiving differe nt P amendments. P, D, and R indicate inorganic P, digested fl ushed dairy manure (FDM), and raw FDM amendments, respectively. Orsino R and Tavares R valu es are absent due to column clogging resulting in discontinued P additions. E rror bars represent standard deviation. Table 38. Duncans Multiple Range Test for Accumulated P. The mean accumulated inorganic P was significantly different than the mean accumulated digested flushed dairy manure (FDM) and raw FDM. The mean accumulated P in the FDM treatments were not significantly different. Clogging of soil columns receiving raw FDM resulted in missi ng data for two soil columns. Duncan grouping Mean (mg P) N P form A 6.0 9 Digested FDM A 5.5 7 Raw FDM B 3.3 9 KH2PO4 Table 3-7. ANOVA output Df Type III SS Mean Square F value Pr > F P source 2 35.34007 17.67003 747.60 < 0.0001 Soil 2 8.57565 4.28782 181.41 < 0.0001 P source soil 4 10.88270 2.72067 115.11 < 0.0001 Rep 2 0.04502 0.02250 0.95 0.4095 40

PAGE 41

Figure 3-7. Correlation between P accumulated in soil columns, delivered as inorganic P at 48 mg L-1 P solution, and soil RPA values. In as cending RPA values, data points include the soils Millhopper E2, Orsino Bw, and Tavares E2. (p< 0.0001) 41

PAGE 42

CHAPTER 4 DISCUSSION Organic matter has been known to promote the leaching of P by competing with or blocking of sorption sites, displa cing P that is sorbed, and medi ating transport of P (Agbenin and Igbokwe, 2006; Andrade et al., 2003; von Wa ndruszka, 2006; Hens and Merckx, 2001). However, as evident in the lack of P breakthr ough in this study among the FDM treated columns (Figures 3-3 and 3-4), other processes must play a major role in the fate and transport of P as applied as FDM on sandy soils in the southeastern United States. Sorption to Fe and Al oxide coatings is a dominant process by which P is fixed in soils similar to these in Florida. Indeed, there was a strong correlati on of the RPA values of the soils used in this experiment and the amount of oxalate extracted Fe and Al they cont ained (Figure 3-1). However, other mechanisms affecting retention of P applied as FDM likely involve components delivered to the soil by the FDM itself, since retention of P applied as FDM did not vary with the so il relative P adsorption capacity as it did for the inorganic P treatment. It is unlikely that the lower DRP concentrations in the FDM could have caused the delayed break throughs shown in Figures 3-3 and 3-4, as the concentrations are not low enough based on an RPA-retardation relation determined for sandy soils by Rhue et al. (2006). Prospective mechanis ms explaining the retention of P added as FDM are (i) precipitation of P with Ca and Mg and (ii) entrapment of particulate P in the soil column. The role of Ca in P fixation in soils is well documented in semi-arid and arid regions where calcareous soils are prevalent (Delgado et al., 2002). This is a major problem concerning farmers, whose fertilizer P b ecomes unavailable for crops. Ma ny authors have written on the subject and have highlight ed the role of organic amendments to increase P availability in the soil (Agbenin and Igbokwe, 2006; Bennani et al., 2005; Fernandez-Perez et al., 2005; Delgado et al., 2002; Cong and Merkx, 2005). This would seem to discourage the use of organic amendments, 42

PAGE 43

such as FDM, in areas sensitive to P leaching, bu t if precipitation of P minerals occurs in the surface horizon, another dynamic process must be considered. Studies have shown that the concentration of Ca in manure is sufficient to induce precipitation of cal cium phosphates (Ca-P), but the minerals are in non-cr ystalline form, which makes them more soluble. (Wang et al., 1995; Hansen and Strawn, 2003). Harris et al. (1 994) and Wang et al. (1995), using x-ray diffraction, were unable to detect any crystalline P minerals in soils that received manure from dairy operations for many years. Organic acids, M g, and Si have all been suggested as preventing crystallization of P minerals (Cooperband and Good, 2002; Harris et al., 1994). Sharpley et al. (2004) determined ion activity products of manu red soils using MINTEQA2 chemical speciation model and ion concentrations in solutions. They reported tricalcium P and octacalcium P as the dominant Ca-P forms in manured soils, and hydroxyapatite as the dominant mineral in nonmanured soils. These minerals are not stable in the manured soil and are subject to leaching (Hansen and Strawn, 2003; Graetz and Nair, 1995.) However, apatite minerals seem to pose relatively little risk of P leaching in soils aff ected by manure as long as soil pH remains in the neutral to alkaline range. Wang et al. (1995) did not find any higher levels of P in leachates derived from soils spiked with apatite than non-spiked soils. If as suspected, Ca-P phases are precipitating, then knowing the sp ecific form will indi cate solubility and release potential. Hansen and Strawn (2003) indicated Ca-P minerals such as octacalcium P and tricalcium P controlled the P concentration of the soil solution in manure amended alkaline soils. Desorption experiments they performed showed a rapid release of P followed by a slow, steady release. This was done using 1g of soil ex tracted with 20 mL of 0.005 M NaCl. After 12 replenishments of such extractions, 29% of th e total P was desorbed from the surface horizon and 8% was desorbed from the subsurface horizon Wang et al. (1995) used simulated rainfall to 43

PAGE 44

leach dairy manure impacted soil samples in soil columns. After 17 weeks of simulated rainfall: each week receiving synthetic rain 80% of the so il pore volume, 13% to 26% of the total P was leached from the most impacted areas. Graetz a nd Nair (1995), after 10 sequential extractions on A and Bh horizons of soils from active dairies found 2% to 18% of the total TP was extracted. The soils, however, kept releasing P at diminished levels in the concluding extractions. They also found higher SRP concentrations in Bh horizons of abandoned dairy site s compared to active sites (43 mg kg-1 to 14 mg kg-1, respectively), supporting the notion that P, over time, will continue to be released and l each. It should also be noted, howe ver, that these sandy soils had little to no P retention capacity, especially the A and E horizons, unlike th e Tavares soil used in this experiment. The change that manure amendments, such as FDM, have on soils is remarkable with regards to P fate and transpor t. As demonstrated in this experiment, P delivered as KH2PO4 solution behaved very differently than P delivered as either digested or undigested FDM in the same soils. The leaching and accumulation of P fr om inorganic P additions could seemingly be predicted from the RPA values of the soils (Figure 3-7). The reported change manure amendments bring about in acid soil s, such as these, is a shift from Al and Fe reaction products to Ca/Mg reaction products, due largely to the great influx of manure-derived bases increasing soil pH (Harris et al. 1994; Sharpl ey et al. 2004). This change is largely confined to the surface horizons as Graetz and Nair (1995) show the shift from Ca/Mg-P predominance in the surface horizon to Al/Fe-P predominance in the deeper horizons and similar nonimpacted soils have Al/Fe-P predominance throughout the profile. Th is is a significant phenomenon because the factors controlling P mobility change as P le aches through the soil. For example, the most important factor organic matter may play in the surface horizon is the inhibition of stable Ca-P 44

PAGE 45

minerals, whereas in the deeper horizons its role of preventing stable Al/Fe-P sorption may be most important. Also, the role of changing pH is important. In the surface horizon the increased pH allows the precipitation of Ca -P minerals leading to their, at least transient, accumulation. In the more acidic subsurface layers, precipitation is not favored. However Harris et al. (1994) found an apatite-like mineral in st ream sediments associated with, but not in, soils that received the manure directly. This may indicate that, in some cases, elements near the P source may inhibit crystallization of apatit e, but elsewhere precipitation of this stable Ca-P may occur. The lack P breakthrough when delivered as FD M shows that, contrary to the original hypotheses, retardation of P movement was not reduced by competition of FDM organic components for P adsorption sites and that mediat ed transfer of P by dissolved organic matter (OM) in the FDM does not play an immediate role in P leaching. Most of the organic P in the FDM is predominant in a particulate form, whic h is prone to entrapment, which would minimize leaching in the short-term, rather than in disso lved form, which could favor reduced adsorption to soil components and greater leaching. The microbe s filtered out of the FDM by the sand in the columns would almost certainly flourish in the column environment. Their growth could retain P and trap it in cellular material. The extent, capacity, and short a nd long term significance of this activity, given the large concen tration of P being added, is unknown, but as evident in the difference in COD between raw and digest ed FDM shown in Table 3-6, the food to microorganism ratio would be higher for the raw FDM. This may also play a factor concerning leaching when FDM is used as fertilizer for crops. The role of organic matter (O M) colloids has been reported as a dominant process in P mobility in high P soils (Motoshita et al., 2003 ). Hens and Merckx (2001) and others have shown that organic matter-metal-orthophosphate (OM-M-P) complexes form in soils. The OM 45

PAGE 46

may be mobile depending on pH, ionic strength, and the ratio of monovalent to divalent cations. They state low ionic strengths and high ratios of monovalent to divalent cations promote organic matter mobility. Also, when they added Al and Fe to soils, the amount of these complexes (defined as high molecular mass molybdate react ive P), increased, but when they just added inorganic P, no increase was obs erved. This indicates that the metal must bond to the OM prior to the P. The soils in their experiment had lo w ionic strengths, low Ca concentrations, and high Na/Ca ratios, whereas the soils in this experiment certainly have the opposite conditions. This may explain why there was little difference in the TP and DRP concentrations in the leachates of this experiment. Hesketh et al. (2001) also investigated colloid transport of P using lysimeters with soils amended with pig slurry and found ve ry little evidence of it; 1% of P applied was leached. The combination of precipitation a nd the maintenance of high divalent cation concentration in the soil may contribute to the prevention of imme diate leaching of P. If FDM amendments are suspended, the pH and amount of di valent cations would d ecrease with rainfall, perhaps increasing the mobili ty of P sorbed to OM and precipitated as Ca-P. An interesting factor in this scenario is the bond strength of the OM-M versus the M-P bond. If OM-M-P complexes leach into E horizons with significant amount s of Al/Fe oxide sand coatings, they could sorb as a solid phase. Hens and Merkx (2001) state that metastable OM-M-P complexes will destabilize with decreasing pH, a condition that would be found in the lower horizons of acidic soils. If the OM-M bond hydrol yzes more easily than the M-P bond, in some instances, P may be occluded without decr easing the sorption cap acity (assuming high concentrations of available P bonding metals). A series of r eactions may occur where P and metals may build upon each other when OM-M bonds destabilize: M-sand + P P-M-sand, 46

PAGE 47

P-M-sand + OM-M OM-M-P-M-sand, OM-M-P-M-sand OM + M-P-M-sand, M-P-M-sand + P P-M-P-M-sand, and so on. The significance of this hypothetical P sink may not be great, but perhaps a study is warranted. This experiment indicates that P entering the soil from FDM application may be in several pools that each has its own release cons tant. About to of the P, from the digested or raw FDM, respectively, is contained in the pa rticulate fraction, which is subject to varying rates of decomposition as soil humus. The remainde r may precipitate with Ca, sorb with Fe and Al oxides or become complexed with OM along with metal ions. In the soils used in this experiment, the FDM immediately creates conditions that result in less P leaching than occurs with inorganic P amendments, which effectivel y rejects the original hypothesis. Indeed, any organic transport is overridden by the accumulati on caused by precipitation and entrapment of P in OM. The rate of FDM addition and the rate of rainfall will be important factors concerning future P leaching. Based on this study and others, P, as delivered by FDM, will likely accumulate to high levels in these soils (Lehmann et al., 2005; Graetz and Nair, 1995). Most of the P will likely be metastable Ca-P along with P associat ed with mineralizable organic particulates, both of which are ultimately subject to leaching under rainfall, particularly when FDM (and hence Ca) additions cease. The complexation of inorganic P with the OM will affect its mobility depending on soil properties down the profile, which can be very different in originally acidic soils. Eventually these soils may leach P at slow, steady rate. The relatively small amount of P leached wh en using digested or raw FDM was very similar between these two treatments. However, less P was removed as DRP for the digested 47

PAGE 48

FDM than for the raw FDM, considering that DRP comprised a higher proportion of TP for the former. The benefit of digested FDM over raw FD M in spray field applications may be more linked with factors other than leaching of P, such as more plant available P and odor and pathogen reduction. The higher proportion of plant available P in the digested FDM would be preferable because more of the P would be available for immediate uptake through the crops when the FDM is applied. These soils have a predictable sorpti on capacity when P is delivered as KH2PO4 and is largely a function of the RPA (Fig 3-7). The soil s can quickly reach their sorption capacity and then offer little resistance to P leaching and leachate P concentrations could be very high. However, the Al/Fe-P is more stable than th e Ca-P minerals formed under the influence of manure (Hansen and Strawn, 2003; Makris et al. 2005). Depending on the environmental, agricultural, and practical conditions, one method of P fertilization may be favored over another. Concerning leaching of P in these soils, when at similar P concentrations, the largest apparent difference between P delivered as FDM and P delivered as KH2PO4 is that when irrigating with FDM, the surface horizon acts as an expanding, but leaky bucket with regards to containing P; that is, the P is accumulated in the surface horizon but steadily leaches at a slow rate. However, the soil acts as a fixed bucket when using an inorganic P fertilizer. P progressively travels down the soil profile becoming more permanently fixed than metastable Ca-P, but reaches its inherent P capacity as Al/Fe-P, and then the so il offers no more resistance to P leaching. As stated above, more analyses of these experime ntal soils could provide useful information concerning P fate and environmental risk. Future Studies Further study of the soil columns could evalua te the evidence of Ca-P precipitation, the extent to which it occurs, and its relevant properties. An analysis such as x-ray diffraction, could 48

PAGE 49

determine if any crystalline precipitant formed, and leaching of the soil columns would be very useful in determining the stability of the accu mulated P. Also, another column study involving soils, with a range of P sorption cap acity (such as those used in th is experiment), overlaid with a highly impacted soil of similar soil type with large amounts of accumulated P may show how P in the leachate is affected by changes in the soil chemistry and if OM leaches into the subsurface layer and prevents sorption at Al/Fe sorption sites. Summary and Conclusions Three different sandy soils in a column e xperiment showed processes related to P leaching differ when delivered as inorganic P as opposed to when delivered by both raw and digested FDM. The capacity of the soils to re tard the movement of P when amended with inorganic P is related to the amount of Fe and Al oxide present in the soil. In contrast, the soils amended with FDM accumulated P and very li ttle leached through the columns. After the addition of 30 pore volumes, the amount of P leac hed from the columns amended with inorganic P was significantly greater than t hose that received both raw and di gested FDM. It is apparent that FDM creates conditions in the soil that prom ote accumulation of P. This is likely due to an increase in pH, from the addition of base cati ons present in the FDM, notably Ca, and a shift from Al/Fe-P to Ca/Mg-P reaction products. Also, any transport of P via OM is minimal compared to the factors promoting P accumulation, and the immobility of OM may also be caused by the changes FDM creates in the soil. The formation of meta-stable Ca-P should be verified by analytical means, and leaching of th e columns would provide data on the stability of the P accumulated in all of the soil columns. Also, as there is a shift back to the original soil chemistry in deeper horizons, leaching of FDM amended soils over the same or similar nonimpacted soils in a column experiment could he lp determine what fact ors are important with relation to P leaching as the soil chemistry changes. 49

PAGE 50

LITERATURE CITED Agbenin, J.O. and S.O. Igbokwe. 2006. Effect of soil-dung manure incubation on the solubililty and retention of applied phosphate by a weathered tropical semi-arid so il. Geoderma 133: 191-203. Anderson, J.M. 1976. An ignition method for determination of total phosphorus in lake sediments. Water Res. 10:329-331. Andrade, F.V., E.S. Mendonca, V.H.V. Alvarez, and R.F. Novais. 2003. Addition of organic and humic acids to latosols and phosphate adsorption effects. Rev. Bras. Cienc. Solo 27: 1003-1011. Akhatar, M.S., B.K Richards, P.A. Medrano, M. DeGroot, and T.S. Steenhuis. 2003. Dissolved phosphorus from undisturbed soil co res: Relation to adsorption strength, flow rate, or soil structure? Soil Sci. Soc. Am. J. 67:458-470. Baril, R. and Bitton, G. 1967. Anomalous values of free iron in some Quebec soils containing magnetite. Can. J. Soil Sci. 47:261. Bennani, F., M. Badraoui, M. Mikou. 2005. Mono calcium phosphate monohydrate concentration in soil suspension amended with orga nic matter. J. Phys. IV 123:159-163. Brady, Nyle C. and R. Weil. The Nature and Properties of Soils. 13th Ed. Upper Saddle River, New Jersey: 2002. Carefoot, J.P. and J.K. Whalen. Phosphorus concen trations in subsurface water as influenced by cropping systems and fertilizer sources. Can. J. Soil Sci. 83:203-212. Chardon, W.J., O. Oenema, P. delCastillo, R. Vriesema, J. Japenga, and D. Blaauw. 1997. Organic phosphorus in solutions and leachates from soils treated with animal slurries. J. Environ. Qual. 26:372-378. Cong, P.T. and R. Merckx. 2005. Improving phosphorus availability in two upland soils of Vietnam using Tithonia diversifolia H. Plant Soil 269:11-23. Cooperband, L.R. and L.W. Good. 2002. Biogenic phos phate minerals in ma nure: implications for phosphorus loss to surface waters Environ. Sci. Technol. 36:5075-5082. Delgado, A., A. Madrid, S. Kassem, L. A ndreu, and M.D. del Campillo. 2002. Phosphorus fertilizer recovery from calcareous soils am ended with humic and fulvic acids. Plant Soil 245:277-286. Dodjic, F., K. Borling, and L. Bergstrom. 2004. Phosphorus leaching in relation to soil type and soil phosphorus content. J. Environ. Qual. 33:678-684. 50

PAGE 51

Eghball, B., G.D. Binford, and D.D. Baltensperger. 1996. Phosphorus movement and adsorption in a soil receiving long -term manure and fertilizer applicatio n. J. Environ. Qual. 25: 1339-1343. Elrashidi, M.A., A.K. Alva, Y.F. Huang, D.V. Calvert, T.A. Obreza, and Z.L. He. 2001. Accumulation and downward transport of phosphorus in Florida soils and relationship to water quality. Comm. Soil Sci. Plant Anal. 32:3099-3119. Fernandez-Perez, M., F. Flores-Cespedes, E. Gonzaalez-Pradas, M.D. Urena-Amate, M. Villafranca-Sanchez, M. Socias-Viciana, and S. Perez-Garcia. 2005. Effect of dissolved organic carbon on phosphate rete ntion on two calcare ous soils. J. Agr. Food Chem. 53: 84-89. Freese, D., S.M. Vanderzee, and W.H. Vanriemsdijk. 1992. Comparison of different models for phosphate sorption as a function of the iron a nd aluminum-oxides of soils. J. Soil Sci. 43:720-738. Geohring, L.D., O.V. McHugh, M.T. Walter, T.S. Steenhuis, M.S. Akhtar, and M.F. Walter. 2001. Phosphorus transport into subsurf ace drains by macropores after manure applications: Implications for best manure management practices. Soil Sci. 166:896-909. Graetz. D.A., and V.D. Nair. 1995. Fate of phosphorus in Florida spodosols contaminated with cattle manure. Ecol. Eng. 5:163-181. Hansen, J.C., T.C. Daniel, A.N. Sharpley, and J.L. Lemunyon. 2002. The fate and transport of phosphorus in agricultural systems. J. Soil Water Conserv. 57:408-417. Hansen, J.C. and D.G. Strawn. 2003. Kinetics of phosphorus release from manure-amended alkaline soil. Soil Sci. 168:869-879. Harris, W.G., H.D. Wang, and K.R. Reddy. 1994. Da iry manure influence on soil and sediment composition: implications for phosphorus retention. J. Environ. Qual. 23:1071-1081. Hens, M. and R. Merckx. 2001. Functional characte rization of colloidal phosphorus species in the soil solution of sandy soils. Environ. Sci. Technol. 35:493-500. Hesketh, N., P.C. Brookes, and T.M. Addiscott. 2001. Effect of suspended soil material and pig slurry on the facilitated transport of pesticid es, phosphate, and bromide in sandy soil. Eur. J. Soil Sci. 52:287-296. Iyamuremye, F., R.P. Dick, and J. Baham. 1996. Organic amendments and phosphorus dynamics. 2. Distribution of soil phosphorus fractions. Soil Sci. 161:436-443. Jiao, Y., W.H. Hendershot, and J.K. Whalen. 2004. Agricultural practices influence dissolved nutrients leaching through intact soil cores. Soil Sci. Soc. Am. J. 68:2058-2068. 51

PAGE 52

Karathanasis, A.D. and D.M.C. Johnson. 2006. Subsurface transport of Cd, Cr, and Mo mediated by biosolid colloids. Sci. Total Environ. 354:157-169. Lehmann, J., Z. Lan, C. Hyland, S. Sato, D. Solomon, and Q.M. Ketterings. 2005. Long-term dynamics of phosphorus forms and retention in manure-amended soils. Environ. Sci. Technol. 39:6672-6680. Leytem, A.B., R.L. Mikkelsen, and J.W. Gilliam. 2002. Sorption of organic phosphorus compounds in Atlantic coastal pl ain soils. Soil Sci. 167:652-658. Lilienfein, J., R.G. Qualls, S.M. Uselman, a nd S.D. Bridgham. 2004. Adsorption of dissolved organic and inorganic phosphorus in soils of a weathering ch ronosequence. Soil Sci. Soc. Am. J. 68:620-628. McKeague, J.A., and D.H. Day. 1966. Dithionite a nd oxalate extractable Fe and Al as aids in differentiating various classes of so ils. Canadian J. Soil. Sci. 46:13-22. Makris, K.C., W.G. Harris, G.A. OConnor, and H. El-Shall. 2005. Long-term phosphorus effects on evolving physicochemical propertie s of iron and aluminum hydroxides. J. Colloid Interf. Sci. 287:552-560. Motoshita, M., T. Komatsu, P. Moldrup, L.W. de Jonge, N. Ozaki, and T. Fukushima. 2003. Soil constituent facilitated transport of phosphor us from a high-P surface soil. Soils and Foundations 43:105-114. Murphy, J., and J.P. Riley. 1962. A modified si ngle solution method for determination of phosphate in natural waters. Anal. Chim. Acta 27:31-36. Nair, V.D., R.R. Villapando, and D.A. Graetz. 1 999. Phosphorus retention capacity of the spodic horizon under varying environmental c onditions. J. Environ. Qual. 28:1308-1313. Novak, J.M. and D.W. Watts. 2004. Increasing the phosphorus sorption capacity of southeastern Coastal Plain soils using water trea tment residuals. Soil Sci. 169:206-214. Pardo, P., J.F. Lopez-Sanchez, and G.Ra uret. 2003. Relationships between phosphorus fractionation and major components in sediments using the SMT harmonized extraction procedure. Anal. Bi oanal. Chem.376:248-254. Phillips, I.R. 2002. Phosphorus sorption and nitrog en transformation in tw o soils treated with piggery wastewater. Aust. J. Soil Res. 40:335-349. Phillips, I.R. 2002. Nutrient losses from undisturbe d soil cores following applications of piggery wastewater. Aust. J. Soil Res. 40:515-532. Qualls, R.G., and B.L. Haines. 1991. Geochemistry of dissolved organic nutrients in water percolating through forest ecosystems. Soil Sci. Soc. Am. J. 55:1112-1123. 52

PAGE 53

Rhue, R.D., W.G. Harris, G.Kidder, R.B. Brow n, and R. Littell. 1994. A soil-based phosphorus retention index for animal waste dispos al on sandy soil. Final Report Florida Dep.Environ. Protection, c ontract number WM459. Rhue, R.D., W.G. Harris, V.D. Nair. 2006. A re tardation-based model for phosphorus transport in sandy soils. Soil Sci. 171:293-304. Sharpley, A.N., R.W. McDowell, and P.J.A. Klei nmann. 2004. Amounts, forms, and solubility of phosphorus in soils receiving manure. Soil Sci. Soc. Am. J. 68:2048-2057. Sharpley, A.N., J.L. Weld, D.B. Beegle, P.J.A. Kleinman, W.J. Gburek, P.A. Moore, and G. Mullins. 2003. Development of phosphorus indices for nutrient management planning strategies in the United States J. Soil Water Conserv. 58:137-152. Siddique, M.T., J.S. Robinson, and B.J. A lloway. 2000. Phosphorus reactions and leaching potential in soils amended with sewa ge sludge. J. Environ. Qual. 29:1931-1938. Slabaugh, J.D., A.O. Jones, W.E. Puckett, a nd J.N. Schuster. 1996. Soil Survey of Levy County, FL. USDA-Natural Resources Conservation Serv ice, National Cooperative Soil Survey. Soil Survey Staff. 1992. Soil Surv ey Laboratory Methods Manual. p. 14. U.S. Department of Agriculture. van Es, H. M., R.R. Schindelbreck, and W.E. J okela. 2004. Effect of ma nure application timing, crop, and soil type on phosphorus l eaching. J. Environ. Qual. 33:1070-1080. von Wandruszka, R. 2006. Phosphorus retention in calcareous soils and the effects of organic matter on its mobility. Geochem. Trans. 7: Art. No. 6. Wang, H.D., W.G. Harris, K.R. Reddy, and E. G. Flaig.1995. Stability of phosphorus forms in dairy-impacted soils under simulated leaching. Ecol. Eng. 5:209-227. Wilkie, A.C. 2005. Anaerobic digestion: biology and benefits. In: Dairy Manure Management: Treatment, Handling, and Community Relations. NRAES-176:63-72. Wilkie, A.C. 2005. Anaerobic digestion of dairy manure: design and process considerations. In: Dairy Manure Management: Treatment, Ha ndling, and Community Relations. NRAES176:301-312. Wilkie, A.C., H.F. Castro, K.R. Cubinski, J.M. Owens, and S.C. Yan. 2004. Fixed-film anaerobic digestion of flushed dairy ma nure after primary treatment: wastewater production and characterization. Bi osystems Engineering 89:457-471. Zhou, M., R.D. Rhue, and W.G. Harris. 1997. Phosphor us sorption characteristics of Bh and Bt horizons from sandy coastal plain soils. Soil Sci. Soc. Am. J. 61:1364-1369. 53

PAGE 54

BIOGRAPHICAL SKETCH Aaron Malek was born in Bryan, TX. He graduated from Navasota High School in 1998 and then received a Bachelor of Science in In terdisciplinary Biology fr om the University of Florida in 2004. He enrolled in grad uate school, also at the Univers ity of Florida, the same year at the School of Natural Resources and the E nvironment. He will graduate in December 2007 and receive a Master of Scien ce in Interdisciplinary Ecology. 54