Citation
Aluminum water treatment residuals for reducing phosphorus loss from manure-impacted, high-watertable soils

Material Information

Title:
Aluminum water treatment residuals for reducing phosphorus loss from manure-impacted, high-watertable soils
Creator:
Rew, Thomas J. ( Dissertant )
Graetz, Donald A. ( Thesis advisor )
Nair, Vimala D. ( Reviewer )
Sollenberger, L. E. ( Reviewer )
Place of Publication:
Gainesville, Fla.
Publisher:
University of Florida
Publication Date:
Copyright Date:
2006
Language:
English

Subjects

Subjects / Keywords:
Aluminum ( jstor )
Boxes ( jstor )
Phosphorus ( jstor )
Rain ( jstor )
Simulations ( jstor )
Soil science ( jstor )
Soils ( jstor )
Subsurface flow ( jstor )
Surface runoff ( jstor )
Water treatment ( jstor )
Dissertations, Academic -- UF -- Soil and Water Science
Soil and Water Science thesis, M.S
Lake Okeechobee ( local )

Notes

Abstract:
Dairy and beef operations in the Lake Okeechobee watershed in Florida and across the nation are receiving attention as a result of their contribution of phosphorus (P) to surficial water bodies. Numerous efforts are being made to support the agricultural industry by reducing P losses from the soil. One such effort involves the addition of water treatment residuals (WTRs) to the soil. Prior research has shown that Al-WTRs are capable of binding P and therefore reducing P loss through runoff and leaching. The objective of this research was to evaluate the effect of Al-WTR on P loss from a manure-impacted soil obtained from a dairy sprayfield using a rainfall simulation protocol. Soil was removed from the field site as 0 10 and 10 20 cm depths. Both depths contained high concentrations of water-soluble P and Mehlich-1 P; approximately 18 and 950 mg P kg-¹, respectively. After air drying and sieving, the soil was placed in rainfall simulation boxes (100 cm x 30 cm x 20 cm) designed to collect runoff, subsurface flow, and leachate. An Al-WTR was either surface applied or incorporated to 10 or 20 cm depths at a rate of 2.5% of soil dry weight. The soil was then sprigged with stargrass (Cynodon nlemfuensis). Rainfall simulations were run six times at 3 wk intervals. Runoff was collected for 30 min after initial runoff began. Subsurface flow and leachate were collected (depths of 10 and 20 cm, respectively) after runoff ceased. When Al-WTR was surface-applied, the SP concentration in runoff was reduced by approximately 75% compared to untreated soil; however, SP concentrations in subsurface flow and leachate did not decrease. When Al-WTR was incorporated into the soil at depths of 0-10 or 0- 20 cm, runoff SP concentrations were reduced by approximately 45%. Incorporation of Al-WTR to a depth of 10 cm decreased SP concentrations in subsurface flow and leachate by 37 and 11%, respectively. However, with incorporation of Al-WTR to a depth of 20 cm, both subsurface flow and leachate SP concentrations were reduced by approximately 90%. The incorporated Al-WTR reduced soil water-extractable P (WEP) by approximately 70%. However, Mehlich-1 P concentrations were not affected by the incorporation of Al-WTR in the soil. Care must be taken to ensure complete incorporation of Al-WTR throughout the P-impacted layer, as Al-WTR is only effective in reducing SP concentrations when it is in contact with the impacted soil. Shoot and root growth of stargrass were not adversely affected by the Al-WTR applied at a rate of 2.5% of soil weight.
Subject:
Al, aluminum, P, phosphorus, rainfall, runoff, simulator, WTR
General Note:
Title from title page of source document.
General Note:
Document formatted into pages; contains 72 pages.
General Note:
Includes vita.
Thesis:
Thesis (M.S.)--University of Florida, 2006.
Bibliography:
Includes bibliographical references.
General Note:
Text (Electronic thesis) in PDF format.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Rew, Thomas 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.
Embargo Date:
3/1/2007
Resource Identifier:
658231283 ( OCLC )

Downloads

This item has the following downloads:


Full Text











ALUMINUM WATER TREATMENT RESIDUALS FOR REDUCINTG PHO SPHORUS
LOSS FROM MANURE-IMPACTED, HIGH-WATERTABLE SOILS














By

THOMAS J. REW


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


2006


































Copyright 2006

by

Thomas J. Rew



































To my Granddad, my partner in crime, for all the summers he encouraged my passion for
agriculture, the environment, and life
















ACKNOWLEDGMENTS

I thank Dr. D.A. Graetz for his guidance and assistance over the past five years

through both my undergraduate and graduate studies. I also would like to thank the other

members of my committee: Drs. V.D. Nair, G.A. O'Connor, and L.E. Sollenberger for

their support and guidance. Special thanks go to the crew who helped collect my soil

sample and prepare the rainfall simulation boxes: Jacob Butler, Eric Smith, Leighton

Walker, and Steve Robinson.

I would also like to thank Dawn Lucas for her help with laboratory experiments,

calculations, and controlling emotions when using the AA; Julie, Lacey, Justin, and Ryan

for their help in the lab (dishes); the CALS Ambassadors and Agronomy and Soils Club

for allowing me to become more than just a student of UF. In addition, I need to thank

the faculty and staff of the University of Florida Soil and Water Science Department

giving me the opportunity to expand my knowledge.

Most importantly, I want to thank my family and friends for the support and advice

they have given me over the past three years. Mom and Dad, I'm finally done; Courtney

Maibach, it is your turn to read my thesis and I promise, I'll finish arguing yours soon.

Finally, Kelly Aynes, thank you for allowing me to setup an office in your house. I could

not have finished without your love, patience, and understanding. I can only hope I can

return love and support over the next two years of vet school. Thank you.



















TABLE OF CONTENTS


page

ACKNOWLEDGMENT S .............. .................... iv


LI ST OF T ABLE S ................. ............... vii........ ....


LIST OF FIGURES .............. .................... ix


AB S TRAC T ......_ ................. ............_........x


CHAPTER


1 INTRODUCTION ................. ...............1.......... ......


2 LITERATURE REVIEW .............. ...............6.....


Forms of Phosphorus in Soil............... ..... .. ............
Water Treatment Residuals as Soil Amendments .............. ..... ............... 1
Aluminum Toxicity .............. ...............14....
Rainfall Simulation............... ...............1


3 MATERIALS AND METHODS .............. ...............20....


Proj ect Description ................. ...... ....... ...............20......
Site Description and Soil Collection ............... ...............20....
Aluminum Water Treatment Residual Collection ................. ............ .........21
Construction of Runoff Boxes ................. ...............21........... ...
Rainfall Simulation............... ...............2

Experimental Design .............. .. ...............26...
Soil Sampling and Forage Harvest ................. ...............27........... ...
Analytical Methods............... ...............27
Soil ................... ... ..... ... .......... .............2
Aluminum Water Treatment Residuals ................ ...............................28

Forage ................ .... ........ ........... .............2
Runoff, Subsurface Flow, Leachate .............. ...............29....
Quality Assurance and Quality Control .............. ...............31....
Statistical Analysis .............. ...............3 1....

4 RE SULT S AND DI SCU SSION ............... ..............3


Initial Soil P Concentrations and pH .............. ...............32....












Aluminum Water Treatment Residual Characteristics ................ ............ .........32
Runoff, Sub surface Flow, and Leachate Characteri stics ................ .....................3 3
V olum e .............. ...............33....

pH .............. ...............34....
Phosphorus ............... ........ ..............3
Nitrate- and Ammonium-Nitrogen ............._... .....__ .......___ ............4
Total Calcium, Magnesium, Iron, and Aluminum Concentrations .....................43
Pre- and Post-simulation Comparisons............... ..............4
Soil Phosphorus ........._ ....... .__ ...............44....
S oil pH ............... ...............45...

Forage .............. ...............46....

5 CONCLUSIONS .............. ...............51....


LIST OF REFERENCES ........._ ....... .__ ...............53....


BIOGRAPHICAL SKETCH .............. ...............61....

















LIST OF TABLES


Table pg

4-1 Initial soil mean values of P concentrations and pH of the Immokalee fine sand
used in the study. .............. ...............32....

4-2 Selected characteristics of the Manatee County Al-WTR. ................ ................. 33

4-3 Average runoff, subsurface flow, and leachate volumes as influenced by
treatment from six simulations of runoff, five simulations of subsurface flow,
and two simulations of leachate. ............. ...............34.....

4-4 Average soluble P concentrations by treatment in runoff(six events), subsurface
flow (five events), and leachate (two events)............... ...............36

4-5 Average soluble P concentrations in runoff, subsurface flow, and leachate after 4
wk of flooded conditions (Simulation 7)............... ...............39...

4-6 Runoff, subsurface flow, and leachate collected during simulation 7(A)
Volume. (B) pH. .............. ...............40....

4-7 Soluble P mass loss (SP concentration volume) per box averaged over six
rainfall events and estimated SP loss per hectare per rainfall event. .......................41

4-8 Total, soluble, and particulate P concentrations in runoff for each treatment
averaged over six rainfall events. .............. ...............42....

4-9 Average N concentrations in runoff, subsurface flow, and leachate from six,
five, and two simulations, respectively. (A) NH4-N. (B) NO3-N. ................... ........43

4-10 Average total Ca, Mg, Fe, and Al concentrations in runoff, subsurface flow, and
leachate from six, five, and two simulations, respectively. (A) Runoff. (B)
Subsurface flow. (C) Leachate. ............. ...............44.....

4-11 Pre- and post-simulation water extractable P concentrations in the 0-10 and 10-
20 cm soil layers............... ...............45.

4-12 Pre- and post-simulation Mehlich-P concentrations in the 0-10 and 10-20 cm
soil layers............... ...............45.










4-13 Pre- and post-simulation soil pH values in the 0-10 and 10-20 cm soil layers........46

4-14 Stargrass shoot biomass harvested prior to each simulation ................. ...............47


















LIST OF FIGURES



Figure pg

3-1 Runoff box design. ........._._. ....._.. ...............22...

3-2 Schematic diagram of rainfall simulator. ......._............_. .......__. .........24

3-3 Box placement under simulator. ...._.._.._ ... .....__. ...._.._ ...........2

3-4 Treatment descriptions .............. ...............26....

4-1 Runoff, subsurface flow, and leachate pH values from six, five, and two
simulations, respectively. ............. ...............35.....

4-2 Average treatment SP concentrations and standard deviations from the three
collection points.. ............ ...............38.....

4-3 Total shoot and root growth per treatment averaged over six simulations. .............47

4-4 Average stargrass TP concentrations per treatment averaged over all
sim ulations. ............. ...............48.....

4-5 Average stargrass TN concentrations per treatment averaged over all
sim ulations. ............. ...............49.....

4-6 Average stargrass Al concentrations per treatment averaged over all
simul nations .. ......._.._.. ...._... ...............50....
















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

ALUMINUM WATER TREATMENT RESIDUALS FOR REDUCINTG PHO SPHORUS
LOSS FROM MANURE-IMPACTED, HIGH-WATERTABLE SOILS

By

Thomas J. Rew

December 2006

Chair: Donald Graetz
Major Department: Soil and Water Science

Dairy and beef operations in the Lake Okeechobee watershed in Florida and across

the nation are receiving attention as a result of their contribution of phosphorus (P) to

surficial water bodies. Numerous efforts are being made to support the agricultural

industry by reducing P losses from the soil. One such effort involves the addition of

water treatment residuals (WTRs) to the soil. Prior research has shown that Al-WTRs are

capable of binding P and therefore reducing P loss through runoff and leaching. The

obj ective of this research was to evaluate the effect of Al-WTR on P loss from a manure-

impacted soil obtained from a dairy sprayfield using a rainfall simulation protocol. Soil

was removed from the field site as 0-10 and 10-20 cm depths. Both depths contained

high concentrations of water-soluble P and Mehlich-1 P; approximately 18 and 950 mg P

kg l, respectively. After air drying and sieving, the soil was placed in rainfall simulation

boxes (100 cm x 30 cm x 20 cm) designed to collect runoff, subsurface flow, and

leachate. An Al-WTR was either surface applied or incorporated to 10 or 20 cm depths









at a rate of 2.5% of soil dry weight. The soil was then sprigged with stargrass (Cynodon

nlemfuensis). Rainfall simulations were run six times at 3 wk intervals. Runoff was

collected for 30 min after initial runoff began. Subsurface flow and leachate were

collected (depths of 10 and 20 cm, respectively) after runoff ceased.

When Al-WTR was surface-applied, the SP concentration in runoff was reduced by

approximately 75% compared to untreated soil; however, SP concentrations in subsurface

flow and leachate did not decrease. When Al-WTR was incorporated into the soil at

depths of 0-10 or 0-20 cm, runoff SP concentrations were reduced by approximately

45%. Incorporation of Al-WTR to a depth of 10 cm decreased SP concentrations in

subsurface flow and leachate by 37 and 11%, respectively. However, with incorporation

of Al-WTR to a depth of 20 cm, both subsurface flow and leachate SP concentrations

were reduced by approximately 90%. The incorporated Al-WTR reduced soil water-

extractable P (WEP) by approximately 70%. However, Mehlich-1 P concentrations were

not affected by the incorporation of Al-WTR in the soil. Care must be taken to ensure

complete incorporation of Al-WTR throughout the P-impacted layer, as Al-WTR is only

effective in reducing SP concentrations when it is in contact with the impacted soil.

Shoot and root growth of stargrass were not adversely affected by the Al-WTR applied at

a rate of 2.5% of soil weight.















CHAPTER 1
INTTRODUCTION

Over the past 5,000 yr, the Lake Okeechobee watershed has developed an

ecosystem to thrive under oligotrophic conditions. The watershed is approximately

12,000 km2 and flows roughly from Orlando to the Everglades (SFWMD, 1997). On

average, the watershed receives 120 cm of rain annually (McCaffery et al., 1976). The

elevation in the watershed is predominately flat, with a range from 3.1 to 22.9 m above

mean sea level (Soil Survey, 2003). As a result, the watertable is naturally close to the

surface for much of the year. Prior to development and utilization of the land, much of

the watershed was completely saturated for one to two months a year (Gatewood and

Bedient, 1975).

The watershed is divided into six regions. The two most important regions are the

Lower Kissimmee River (LKR) and Taylor Creek/Nubbin Slough (TCNS) because they

contribute approximately 57% of the total phosphorus (P) load to Lake Okeechobee

(Fluck et al., 1992). Phosphorus discharge to surficial water bodies has recently become

a maj or concern in the watershed. One of the current maj or contributors to the situation

is waste from livestock which contains high amounts of P. As Florida has continued to

develop, land has become more valuable and farmers and ranchers have intensified their

operations by increasing the concentration of animals per acre. The resulting effect has

been increased discharge of P into the watershed.

Because of the warm climate and abundance of water, the agricultural community

has flourished, and land use practices in the LKR and the TCNS have intensified to









maximize production. In particular, the beef and dairy industries have, respectively,

amassed herds of 133,000 and 32,000 animals in Okeechobee County (Florida

Agriculture Statistical Directory, 2004). To support these herds, 31% of the land in the

LKR and 62% of the land area in the TCNS has been drained and converted to improved

pastures. Improved pasture area increased by 500% between 1950 and 1970. Currently,

improved pastures comprise approximately 87,000 ha in the LKR and the TCNS (Flaig

and Havens, 1995). Inorganic fertilizer used on improved pastures account for 34% of all

P imports to the watershed (Boggess et al., 1995). Additional P imports come from

mineral supplements and winter feed. The most intensive land use in the watershed is

dairy farming. From 1960 to 1990, dairies increased their imports of P in cattle feed from

360 t P yr- to 1200 t P yr- (Flaig and Havens, 1995; Boggess et al., 1995). Dairy feeds

constituent 35% of P imports into the watershed, but P impacts have been decreasing

since 1990 as a result of the dairy buyout program which closed 19 of 49 dairies in the

basin (Flaig and Reddy, 1995).

Phosphorus is a vital nutrient in both natural and agricultural environments and is

often imported to foster agricultural operations. Traditionally, P has been considered

relatively immobile in soils, primarily only subject to loss by erosion. However, P can

also move vertically or horizontally within some soil profiles and can be transported into

the surface water of canals, streams, and lakes. Graetz and Nair (1995) found residual P

from manure has the potential to leach laterally because of the surface horizon' s low P-

retention capacity in a Spodosol. The resulting soluble P in surface water often causes

detrimental impacts on native ecosystems.










Phosphorus occurs in soluble and insoluble compounds in both organic and

inorganic forms in the soil environment (Faulkner and Richardson, 1989). Organic forms

exist as both living plant tissue and in deposits of detrital material. Inorganic P is found

as insoluble phosphate and minerals adsorbed to clay particles, organic matter, and Fe

and Al hydroxides (Mitsch and Gosselink, 1993). Nevertheless, the majority of P is

unavailable to biota. Unlike nitrogen, P does not have any significant losses to the

atmosphere. Therefore, P remains in the soil or is transported by another mechanism

besides volatilization.

Soils in the watershed consist predominately of Spodosols, Entisols, and Histosols.

The Histosols occur primarily as small deposits in wetlands, where as Spodosols

dominate the northern portion of the watershed. The surface horizons are typically

between 0.08 and 0.20 m thick and are underlain by a spodic horizon at 0.5 to greater

than 2 m depth (USDA, 1990). The closer the spodic horizon is to the surface, the more

poorly drained the soil. Spodosols are characterized by high infiltration rates because of

the greater than 90% sand content. However, they have poor internal drainage because of

the low permeability of the spodic horizon. The Spodosols, Myakka (32%), Immokalee

(30%), and Pomello (2%) cover 64% of Okeechobee County (Soil Survey Staff, 2003).

The concern regarding the high watertable and high infiltration rates is that surface

horizons of Entisols and Spodosols have a limited capacity to retain P and, thus, are

susceptible to P loss.

The Bh (spodic horizon) in South Florida Spodosols contains Al and/or Fe and

organic material accumulated from the stripping of the E horizon. Spodosols in general

are naturally deficient in P (Hodges et al., 1967). Nevertheless, as a result of the









accumulation of material, the spodic has the ability to retain downward moving P onto Fe

and Al oxides (Nair et al., 1995). These eluvial materials enable the possible retention of

excess P from agricultural operations. Depth to Bh has an impact on the P leaching

potential of the soil. As the depth of the Bh increases the potential for P loss via

subsurface drainage increases. Moreover, as the depth of the spodic increases its P

adsorption capacity decreases (Nair et al., 1999).

Despite the sorption of P in the soil, it can still be highly mobile. The combination

of the fluctuating and perched watertable allows lateral movement of P through the soil

and into the surface water (Graetz and Nair, 1995). In fact P can move so readily, that

between 1973 and 1988, the concentration of P in Lake Okeechobee increased

approximately 250% (Neganban, 1993).

In an effort to reduce eutrophication and restore the Everglades to a system

resembling the past, larger releases of water to the Everglades and reductions in P loading

are mandatory. Therefore, to meet the requirements of the Comprehensive Everglades

Restoration Plan, the South Florida Water Management District, U.S. Army Corps of

Engineers, and other state agencies created massive water treatment areas, instituted use

of Best Management Practices (BMPs), and increased water flow to the Everglades

(SFWM, 1997). Representative BMPs include fencing off canals, adding water troughs

to pastures, and reducing P concentrations in feed.

These measures alone are not adequate to meet the acceptable P levels within the

watershed due to residual P levels remaining in soils highly impacted by manure. Recent

studies have suggested that the addition of soil amendments has the potential to reduce P

loss from soil. Investigations have included numerous amendments from differing









sources. One promising area of investigation is the land application of water treatment

residuals (WTRs). WTRs are the waste product from drinking water purification, and

typically contain Al, Ca, or Fe compounds used to remove nutrients and particulate

matter from source water. Land application of WTRs has the potential to help reduce

water soluble P (WSP) concentrations in P-impacted soils. Previous research has shown

that WTRs containing Al are the most efficient in reducing WSP in soils. Therefore, a

study was developed to evaluate the effect of an Al-WTR on soil P dynamics in a sandy

soil with a fluctuating watertable. The study had two main objectives:

1. To determine the effect of surface application, partial mixing, and complete mixing
of Al-WTR with soil on P loss in runoff, subsurface flow, and leachate.

2. To determine the Al-WTR' s effect on forage growth.















CHAPTER 2
LITERATURE REVIEW

Forms of Phosphorus in Soil

Phosphorus (P) is commonly found in soils in small total quantities (Lindsay,

1989). Forms of P in soils can be either inorganic or organic. Phosphate minerals,

organometallic complexes, and P to bound metals and hydroxides are all inorganic forms

of P that may be present in the soils. Microbes, humus, and undecomposed litter make up

the organic P fraction (Gale et al., 1994). Reddy et al. (1996) identified five major pools

of P in soils: a) labile inorganic P, b) P bound to Fe and Al minerals, c) P bound to Ca

and Mg minerals, d) P bound in labile organic forms, and e) residual organic P.

The labile pool of P is of particular environmental concern because labile P can

potentially move laterally or vertically in the soil profile as a result of precipitation. The

Glossar of Soil Science Terms (SSSA, 2006) defines the labile pool of P as "that portion

which is readily solubilized or exchanged when the soil is equilibrated with a salt

solution," and the available pool as "the amount of soil P in chemical forms accessible to

plant roots or compounds likely to be convertible to such forms during the growing

season."

Oxalate-extractable Al and Fe dominate P retention in many soils and provide

strong sorption of P (Ballard and Fiskell, 1974). The retention of P by Fe/Al or Ca/Mg

depends on the soil's pH. In acid and neutral soils, P retention is dominated by Fe and Al

compounds because P binds with protonated surfaces of hydroxides of Fe and Al (Olila

and Reddy, 1995). In alkaline soils, Ca and Mg compounds are more dominant in









retaining P because P sorption decreases due to competition between PO43- and OH-

anions. Anion exchange binds phosphates with Fe and Al hydroxides (Olila and Reddy,

1995). Because amorphous Al and Fe hydroxides have a larger number of singly

coordinated surface hydroxyl ions, they have a higher P sorption capacity than crystalline

oxides (Reddy and Smith, 1987). In reduced conditions, Fe stability is affected. As a

result of the reduced stability, P associated with Fe3+ maybe solubilized and released into

the soil solution. However, Al compounds associated with P are not affected by changing

oxidation and reduction conditions in the soil (Miner, 2001).

Labile organic represent the short term storage of P as living tissue in plants. This

P is quickly returned to the soil either through manure or decomposition, whereas,

residual organic represent long term storage of P as detrital tissue that is resistant to

decomposition (Reddy et al., 1996). Because of the P storage ability of plant tissue,

phytoremediation of warm-season perennial grasses has been examined. Warm-season

perennials have the ability to extract more P than warm-season annuals and cool-season

grasses because the potential dry matter yield of perennials is greater. Howard (2006)

estimates that adding nitrogen (N) to increase forage yields in hay production could

decrease soil P concentration 4 to 13 mg kg-l annually.

The soil's ability to retain P determines the risk of environmental consequences

related to P leaching. Therefore, the continued assessment of a soil's capacity to retain P

remains a vital tool in environmental sustainability. Soluble forms of P react and form

less soluble compounds with soil components. The amount of P available to be removed

from solution depends on the quantity of P-reactive colloidal surfaces within the soil

(Holford et al., 1997). The sorption and desorption capacity of soils between horizons









has the potential to influence the movement of P within the profile (He et al., 1999).

Variations of P sorption are attributed to surface area variability based on particle size

(Atalay, 2001). Soil sorption and desorption mechanisms control the solution phase P

(Frossard et al., 1995).

Determining the movement and availability of P in soils and groundwater is

paramount in the preservation of natural resources. By using the Langmuir adsorption

isotherm equation, an estimate of P sorption capacity and strength can be determined in

the soil. Soil extractable P can be related to the P-sorbing capacity of a soil by using the

degree of P saturation test (DPS); DPS = extractablee soil P / P sorption maximum) 100

(Breeuwsma and Silva, 1992). Degree of P saturation may also be expressed as a

percentage of double-acid extractable P (Mehlich-1) to the P sorption capacity of a soil

(Nair and Graetz, 2002).

When sorption equals desorption a system is at equilibrium. Therefore, if the

concentration of P in soil water decreases, desorption will occur until equilibrium is

reached. Likewise, if the P in soil water increases, sorption will occur until equilibrium is

attained (Froelich, 1988). The occurrence of no net sorption or desorption is known as

the equilibrium P concentration (EPC). Sharpley and Menzel (1987) found sorption

properties of the soil control the conversion from dissolved P to particulate P and back to

dissolved P. Therefore, the concentration of P controls the direction of exchange. A

prediction of loss or gain of P from solution can be determined using the EPC; soils with

low EPC values tend to sorb soluble P.

Graetz and Nair (1995) found that the A and E horizons of a manure-impacted

Spodosol had higher EPC values than the Bh horizon. Therefore, A and E horizons are









inferred to have a lower retention capacity than the Bh horizon. In manure-impacted

soils, high EPC values reflect high P loading. Graetz and Nair (1995) also found that

oxalate-extractable Al and total organic carbon accounted for approximately 69% of the

variability in the P retention maximum in the A, E, and Bh horizons forage, pasture, and

intensive areas. Maximum P retention capacity increases with depth. In fact, the Bh was

found to have a P retention capacity three to four times greater than the A and E horizons

(Reddy et al., 1996).

The high cost of inorganic N fertilization has prompted agriculture to better utilize

waste products through the application of biosolids and animal manures. These

amendments are typically applied to meet a crop's N requirement. The result of this

common practice is an excessive accumulation of P in the soil. Soils with low P sorption

capacities, such as sandy soils, are more susceptible to P loss through runoff and

subsurface water movement. Kleinman and Sharpley (2003) evaluated P runoff from two

soils amended with three types of manure (applied at 6 rates ranging from 0-150 kg TP

ha l) following the National P Protocol for rainfall simulation studies. Dissolved reactive

P (DRP) in runoff increased with increased manure application rate. Water-extractable P

(WEP) concentrations in the manures were directly related to DRP runoff concentrations.

Repeated rainfall events diminished DRP concentrations in runoff with all manure types

and application rates. This trend was attributed to both the translocation of manure P into

the profile and the loss of previous applied P by runoff.

A close association between soluble P (SP) concentration in land-applied manures

and P concentrations in runoff has been shown in recent studies. Moore et al. (1994) and

Withers et al. (2001) indicated that DRP loss from amended soils was proportional to the









soluble P content of the source. In addition, Kleinman et al. (2002) found the WSP

concentration of surface applied manure (dairy, swine, and poultry) was highly correlated

with DRP in runoff from three soils.

Water Treatment Residuals as Soil Amendments

Efforts to reduce P loss in agricultural sandy soils include several strategies:

reducing P loading, increasing the ambient P storage, and maximizing P retention. Many

soils with high P concentrations currently discharge P into surfieial water bodies.

Phosphorus retention in highly impacted sandy soils could be improved with the addition

of amendments such as water treatment residuals (WTRs) derived from the treatment of

drinking water to remove color, taste, turbidity, and odor. The chemical composition of

WTRs depends on the metal salts used to clarify the water. The typical salts used are Al

(alum), Fe (ferric chloride), or Ca (CaCO3). The by-product of this water purification

process is solid material commonly referred to as WTR. The WTR is often considered a

waste product because it is currently being disposed in landfills, sanitary sewers, or in

lagoons (Ippolito et al., 2002). Water treatment residuals have historically had little

success as a soil substitute or soil amendment compared to their counterpart, biosolids

(Cornell and Westerhoff, 1981). However, recently WTRs have been considered as soil

amendments because of the P-sorbing characteristics.

Recently WTRs have been evaluated for use as both a soil substitute and a soil

amendment because their properties are similar to Eine textured soils (Elliott et al., 1990).

These Eine textured substances are comprised of sand, silt, and clay particles as well as

activated carbon, polymers, Al, Fe oxides, and calcium carbonate derived from the raw

water (Elliott and Dempsey, 1991). The WTRs have improved soil conditions in

numerous studies by increasing organic matter content, water holding capacity, and pH.









However, because of the high relative surface area of the amorphous Al and Fe

compounds, Al- and Fe-WTRs can cause P deficiencies in soils and reduce crop yields

(Bugsbee and Frink, 1985).

Dayton and Basta (2001) tested the beneficial properties of WTRs by evaluating 17

WTRs for use as a soil substitute by comparing their nutritive, physical, and chemical

properties with soil. The WTRs contained the full spectrum of available nutrients, except

P. A bioassay was performed with tomato seedlings (Lycopersicon esculentum) to

validate the results of soil tests used to measure P adequacy in the WTRs. All plants had

low tissue P (561-1840 mg kg- ; median 923 mg kg- ); 1000 mg kg-l is considered

defieient. Vegetative yield was limited primarily because of P deficiencies and in some

cases, phytotoxic levels of NO2-N (>10 mg kg- ).

In an effort to utilize the positive aspects of organic wastes and minimize

detrimental effects, Gallimore et al. (1999) examined four surface application treatments

of Al-WTR (0, 1 1.2, 44.9 Mg ha-l and 44.8 Mg ha-l in a buffer strip at the end of the plot)

on bermudagrass plots receiving 6.72 Mg ha-l of poultry litter. The plots received

simulated rainfall for 75 mins at 6.3 cm hr- within 24 hr of the application of the litter

and Al-WTR. Surfieial runoff was collected and analyzed for TN, NH4-N, TP, SP,

dissolved Al, and dissolved solids. No reductions in SP concentrations were observed

with the 11.2 Mg hal rate. Soluble P concentration was reduced from 15.0 mg L^1 in the

control to 8.6 mg L^1 in the 44.9 Mg ha-l application rate. Similar results were obtained

when the WTR was applied just in the buffer strip. These reductions were attributed to

the amorphous Al in the Al-WTR. In addition, dissolved solids and Al content did not

increase in surface runoff with the application of the Al-WTR.









The potential P-sorption and precipitation capability of Al-, Ca-, and Fe-WTRs has

generated interest in their use as an amendment in highly P-impacted soils. O'Connor et

al. (2002) found that WTRs reduced P solubility and leaching in manure-amended soils.

Brown and Sartain (2000) reported that Fe-WTR was able to significantly reduce P

leaching on a simulated golf course using 2.5% by weight of the Fe-WTR. Al-WTRs

reduced P and NH4' by 75% in surface runoff from land treated with poultry manure

(Basta and Sloan, 1999). All soluble P was adsorbed in an 8:1 mixture of Al-WTR and

biosolids (Fort Collins, CO). Increasing this ratio has the potential to adsorb all available

P in the biosolids and the soil P (Ippolito et al., 2002).

A rainfall simulation runoff study by Haustein et al. (2000) evaluated both Al-WTR

and HiClay Alumina (HCA-a byproduct of commercial alum production) to test P

adsorption capabilities in a soil highly impacted by P. The HCA P-adsorption capacity

was 20 times less than the Al-WTR P-adsorption capacity (86 vs. 1750 mmol kg- ). In

addition, the Al-WTR increased the total recoverable Al in the soil, while HCA did not

affect the recoverable Al concentration. These differences were attributed to the greater

total Al content of the Al-WTR compared to the HCA (159 vs. 46.7 g kg- ) .

The effect of WTR composition on P solubility and leaching was tested by Elliott

et al. (2002) in a sandy soil (Immokalee-Spodosol) after additions of biosolids and triple

superphosphate (TSP) to increase P concentrations in the soil. The study evaluated the

ability of Al-WTR, Fe-WTR, Ca-WTR, and pure hematite to reduce P loss from the soil.

Soluble P concentrations in the leachate decreased in the order: Al-WTR > Ca-WTR -

Fe-WTR>>hematite.









The retention mechanisms of an Al-WTR were studied by Ippolito et al. (2003).

The research involved shaking an Al-WTR for 1-211 d and analyzing the solution for

pH, Ca, Al, and P. The pH increased from 7.2 to 8.2 after shaking. Change in pH

beyond 84 d of shaking was not significant. In addition, the shaking increased Ca and Al

concentration and decreased P concentration. The average maximum pH of 8.15 and Ca

desorption suggest that the Al-WTR' s Ca source was the raw water used at the treatment

plant (South Platte River pH~7.5-8 and Ca = 41.9 g kg- ). Removal of P was attributed

primarily to chemisorption on the amorphous Al mineral phase.

Novak and Watts (2004) evaluated the impact of incorporating two Al-WTRs on

two Ultisols in a laboratory setting. The Al-WTRs were added at the following rates:

2.5, 5.0, 7.5, and 10.5% by weight. The Pmax values for both amended soils (175 and 85

mg P g- ) were significantly higher than Pmax values without the addition of Al-WTRs

(<1.0 mg P g- ). The results demonstrate the usability of Al-WTRs to increase P

adsorption in sandy soils. In addition, the findings suggest off-site P-transport could be

reduced by the implementation of this new chemical based BMP.

Application rate of an Al-WTR (0, 2.5, 5.0, and 10.0% by weight), the degree of

mixing of the Al-WTR with the soil, and the mixing of an un-impacted E-horizon with an

impacted A-horizon on P leaching was examined by Miyittah-Kporgbe (2004). Runoff

and leaching rates were determined via rainfall simulations and columns, respectively

(leachate was also collected from rainfall simulation.). Phosphorus leaching was reduced

by 87 to 99.7% when the Al-WTR was mixed completely with 15 cm of soil, compared

to partial mixing (7.5 cm of soil), which only reduced P leaching from 40 to 58%.

Improved P retention was attributed to increased contact of the Al-WTR with soil P. The









runoff results found that soluble P concentrations met the critical threshold of <0.03 mg

L when Al-WTR was surface applied. However, due to limited contact between the

amendment and soluble P, the leachate soluble P concentrations exceeded the critical

value. Combining the impacted A-horizon with the un-impacted E-horizon increased the

Al-WTR efficiency. This increase in efficiency was attributed to the dilution of soluble

organic material, which can block P adsorption sites on the Al-WTR. Conclusions from

the data suggest Al-WTR must be in direct contact with soluble P to prevent loss by

leaching. In addition, an increase in surface application (10%) can be as effective as

mixing lower quantities (2.5%) of Al-WTR. Application and distribution of Al-WTR in

the soil must be made in accordance with anticipated P losses; surface application with

runoff or mixing with leaching.

Aluminum Toxicity

Aluminum toxicity to plants is a primary concerns when Al-WTR is applied to soil.

In acidic soils, Al toxicity has been shown to be one of the primary growth-limiting

factors (Foy et al., 1978). Mobility and phytotoxicity of Al in the soil solution as a result

of soil acidiaication (Taylor et al., 1989) is a potential problem. Soil acidity impacts

approximately 40% of the arable soil in the world (Haug, 1984). As a result, Al

phytotoxicity presents a serious agricultural concern (Van Wambeke, 1976). Symptoms

of Al toxicity, such as reduced root and shoot growth, mimic phosphate, Ca, and Fe

deficiencies (Foy, 1984). Extensive Al research has been conducted which indicates Al

toxicity primarily affects the roots: a) the observable symptoms of Al toxicity are the

reductions in root elongation (Osborne et al., 1981; Jarvis and Hatch, 1986); b) the

production of root biomass compared to shoot biomass is typically more sensitive to Al









(Buss et al., 1975; Zhang and Taylor 1988); c) in Al-stressed plants there is no correlation

between an accumulation of Al in the shoots and the supply of Al (Foy et al., 1972); and

d) the Al tolerance of a cultivar has been shown to be determined by the rootstock in

graftmng experiments (Klimashe, 1970).

The visual appearance of mature roots is not the exclusive determination of Al

toxicity or tolerance. At the cellular level, Al influences leaf tissue metabolism (Ohki,

1986; Hoddinott and Richler, 1987; Sarkunan et al., 1984). Taylor et al. (1989)

suggested potential mechanisms of Al disruption on cell functions to be: a) disruption of

membrane structure and functions; b) inhibition of DNA synthesis and mitosis; c)

inhibition of cell elongation; d) disruption of mineral nutrition; and e) disruption of

phosphate and Ca metabolism. Taylor also suggested several possible immediate toxic

effects of Al. The membrane structure and/or function at the soil root interface maybe

altered. Membrane-bound enzymes will be affected. Aluminum may also reduce cell

elongation by affecting cell wall components or assembly. Phosphate in the DNA

appears to bind with Al in the cytosol. As a result, DNA synthesis will be inhibited

because of repressed template activity. Finally, toxic effects of enzyme-mediated

reactions in phosphate metabolism seem probable.

Aluminum occupies approximately 7% of the of earth' s crust. Based on the

toxicity issues described previously, it is fortunate the availability of Al is reduced by

ligands or the Al occurs in other nonphytotoxic forms (i.e., aluminosilicates and

precipitates) (Delhaize and Ryan, 1995). However, the solubilization of Al is enhanced

by low pH. Trivalent Al (Al3+) dominates in acidic conditions (pH<5), compared to

Al(OH)2+ and Al(OH)2+ that form as pH increases. Near neutral pH, the solid phase









gibbsite occurs (Al(OH)3) while in alkaline conditions, aluminate (Al(OH)4~) forms.

Many of the monomeric Al cations have the ability to bind with organic and inorganic

ligands, including: PO43-, SO42-, F-, organic acids, proteins, and lipids (Delhaize and

Ryan, 1995).

Since Al3+ is predominately restricted to acidic conditions and many trivalent

cations are toxic to plants, it is typically assumed to be the maj or phytotoxic species.

However, based on the complex nature of Al, this has been difficult to prove conclusively

(Delhaize, 1995). In addition, most of the monomeric Al species previously listed have

been considered toxic in one or more studies (Kinraide, 1991). To combat toxicity,

numerous strategies have been utilized. The method preferred in North America and

Europe has been the application of calcium carbonate (lime) to raise soil pH. As the pH

rises, the Al is converted to less toxic forms (Samac, 2003).

Aluminum toxicity and heavy metal contamination concerns have been raised

regarding the land application of Al-WTRs. Dayton and Basta (2001) tested 17 WTRs

(14 were alum based) for heavy metals and nutrients according to the U.S. Environmental

Protection Agency's (EPA) toxicity characteristic leaching procedure (TCLP), a test

designed to determine the mobility of organic and inorganic analytes in liquids, solids,

and multiphasic wastes (U. S. EPA, 1988). The results of the test found that all of the

WTRs contained significantly lower than regulatory levels of the EPA' s TCLP and the

residuals were therefore, classified as nonhazardous wastes. Total N ranged 1.3 to 18.4 g

kg-l with an average value of 7 g kg- Typical soil levels of soil total N range from 0.2 to

5.0 g kg-l (Dayton and Basta, 2001). The sources of higher N levels in the WTRs are the

algae, detritus, etc. removed from raw water. Mineralization of organic N in the WTRs









may pose potential problems related to NO2~ COncentration in WTRs containing higher

than 10 g kg-l total N. Most significantly, Dayton and Basta (2001) found soluble Al

levels at a median concentration of 0.054 mg L^1, with a range from 0.02 to 0.92 mg L^1

Therefore, problems stemming from both Al toxicity and heavy metals are not expected

with WTR use. Elliot and Depsey (1991) found that unless the raw water source is

contaminated, the nutrient content of WTRs is generally low. As a result, the WTRs pose

only a minor threat to the environment.

Rainfall Simulation

Rainfall simulation studies over the past 40 yr have become popular because

irregular distribution of rainfall hampers the possibility of reasonable time periods for

study (Neff, 1979). Simulators were first used for erosion studies. The two primary

issues were raindrop size distribution and energy (Esteves, 2000). Through a variety of

research, two types of rainfall simulators have emerged: (i) drip former (Farmer, 1973;

Romkens et al., 1975; Munn and Huntington, 1976) and (ii) nozzles (Meyer and McCune,

1958; Swanson, 1965; Miller, 1987; Riley and Hancock, 1997). Pressurized nozzle

systems have become the preferred method for large area field studies (10 to 500 m2)

(Esteves et al., 2000). Simplicity and speed are the basic requirements for the

movement and assembly of a rainfall simulator from one research location to another. As

a result, compromises have been made between technical constraints and the reproduction

of natural rainfall characteristics (Esteves et al., 2000).

A critical factor for experiments using rainfall simulators is the estimation of

rainfall at the ground level (Yu et al., 2003). As rain gauge's catchment area is elevated

above the ground's surface, the pressure of simulated rain from the nozzle decreases as

greater distances are reached. The elevated catchment area has the possibility to









overestimate rainfall volume. Overestimation without adjustment could range from six to

29 percent depending on catchment height (Yu et al., 2003). Additional variability can

occur due to water pressure differences, nozzle aging, and imperfect nozzle spray

overlap. Wind can also cause non-uniform spatial variation. Measures to avoid

variability include better pressure control, stainless steel nozzles, and the use of a

windbreak (Yu et al., 2003; Esteves et al., 2000).

An additional challenge for research is the comparability of experimental results

from different researchers. The potential for human error is enhanced by differing

sampling techniques, recording methods, measurements, and setup. To avoid differences

in results, the National Phosphate Research Proj ect (NPRP, 2001) adapted a specific

design for rainfall simulators. The establishment of a single design for a simulator

expedites data collection, promotes comparable results, and attempts to maintain field

relevancy (Sharpley and Kleinman, 2003). The designs of Shelton et al. (1985) and

Miller (1987) are the basis of the portable rainfall simulator used by the NPRP (Humphry

et al., 2002).

Numerous studies across the nation have used the NPRP protocol for rainfall

simulations both for field and box studies to evaluate P loss under differing soils and

amendment treatments. Kleinman et al. (2003), Moore et al. (2000), and Withers et al.

(2001) have used rainfall simulators to demonstrate P loss in runoff. As our knowledge

base has increased with these studies so has the need with the design of the rainfall boxes.

Miyittah-Kporgbe (2004) has used NPRP's runoff box design and added a false bottom to

collect leachate and examine WTR' s effect on both runoff and leachate. Further studies






19


are needed to investigate the impact on soluble P with the mixing of WTRs at differing

depths and the impacts on plant growth and chemical composition.















CHAPTER 3
MATERIALS AND METHODS

Project Description

Site Description and Soil Collection

Soil for this study was obtained from a tile-drained sprayfield at Larson Dairy Barn

5 near Okeechobee, Florida (N 27016.088', W 80046.460'). The Hield had received

applications of dairy manure for several years. In addition, effluent from the farm's

lagoon was applied to the field periodically as needed to maintain lagoon capacity. The

soil was an Immokalee Eine sand (sandy, siliceous, hyperthermic Arenic Alaquods)

brought into agricultural production as a pasture. The site was planted with stargrass

(Cynodon nlemfuensis) for silage production in 1990. A series of surface soil samples

was taken over the entire Hield to select a sample site with relatively high P levels. These

samples were analyzed for water-soluble P using a HACH Orthophosphate Test Kit and

areas of high P levels were identified. Based on these results, an area within the field was

selected and a more intensive grid sampling was conducted to identify the final sampling

site. These samples were then analyzed for water soluble P (WSP) using a 1:10 soil to

solution ratio and Mehlich-1 extractant (Mehlich, 1953) to determine P concentration.

Bulk soil samples were obtained from 0-10 cm and 10-20 cm depths at the

selected site in a 6.5 x 1.5 m area after removal of above-ground vegetation. The soil was

transported to Gainesville, FL, air dried, and sieved (0.64 cm mesh) to remove debris and

to homogenize the soil. Five samples from the bulk dried and sieved soil were randomly









collected per depth (10 in total) and analyzed for WSP and Mehlich-1 P to evaluate

uniformity of the soils prior to packing the rainfall simulator boxes.

Aluminum Water Treatment Residual Collection

The aluminum water treatment residual (Al-WTR) for the study was obtained from

the Manatee County water treatment plant in Bradenton, FL. Slurry ponds were used to

hold the Al-WTR until it was dry enough to be moved with a front-end loader. The solid

material was then moved to spoil piles to continue to dry before it was disposed of at a

landfill. The Al-WTR samples used for this study were collected from the spoil piles.

Size distribution of the Al-WTR particles varied widely. Material ranged from clay sized

particles to 10 cm fragments. The Al-WTR was air dried for 2 wk under an open-sided

greenhouse, because of its high moisture content, before sieving to pass a 0.64 cm mesh.

Even after the material was dried and sieved, the moisture content was 29% because of

the high organic content of the Al-WTR. Five samples were randomly collected from the

bulk Al-WTR and analyzed for WSP, Mehlich-1 P, and TP.

Construction of Runoff Boxes

Runoff boxes (100 cm long by 30 cm wide and 20 cm deep) were constructed using

pine lumber (Fig. 3-1). A 4 cm rail was added to raise the bottom of the runoff box. A 2

cm lip surrounded the box on three sides to prevent loss of water through runoff and/or

splashing. One end of the box was 2 cm lower to allow runoff to be collected. The box

corners and sampling ports were sealed using latex caulk, and the entire box was covered

with a gel coat to make the box watertight. Ports used to collect subsurface flow and

leachate were covered with a double layer of garden weed screen to prevent blockage of

the ports with debris. Thirty-nine kg of the sieved 10-20 cm depth and then 39 kg of 0-

10 cm depth were added to replicate field soil depths.










A slit PVC pipe (5.6 cm ID by 40.6 cm length) with end caps was attached to one

end of the box to collect runoff. Gravity allowed the runoff water to be collected through

a port in the center of the pipe. Six drainage ports with attached Tygon tubing were

installed on the bottom of the boxes to monitor the watertable and to collect leachate (Fig.

3-1). The ports were located in pairs: 77.0, 38.5, and 5.6 cm from the runoff collection

end and 5.6 cm from the side walls. In addition, two ports with attached Tygon tubing

were placed at the end of the box between the two soil layers to collect subsurface flow.












3 3

3py VIidBAK V~FIEW


~IIUI~ Y IL_


1. Runoff collection point.
2. Subsurface drainage points.
3. Leachate drainage points.


Figure 3-1. Runoff box design.









Soil was placed in the boxes either as discrete 0-10 cm (39.5 kg) and 10-20 cm

(39.5 kg) depths or as a mixed layer at approximately 1.3 g cm-3 to mimic field conditions

as noted in the experimental design section below. Al-WTR from Manatee County was

applied to the appropriate treatments at a rate of 2.5% of dry soil weight (0.975 kg of Al-

WTR dry weight per 10 cm of soil treated). The Al-WTR and soil were mixed using a

cement mixer to evenly distribute the Al-WTR and/or soil in the appropriate boxes. After

establishing the appropriate treatments, stargrass cuttings obtained from an area adj acent

to the sampling site in the sprayfield were sprigged at 30 per box to simulate Hield density

of grass. Simulation boxes were kept in an open-sided greenhouse during the study to

protect them from rainfall.

Rainfall Simulation

Rainfall simulation was conducted according to a protocol developed by the

National Phosphorus Research Proj ect (NPRP) to quantify soil P-runoff relationships

(NPRP, 2001). Rainfall intensity was uniformly applied at 7.1 cm h-l (approximately 210

cm sec l) from a height of 3 m above the soil surface. This is equivalent to a 10 yr, 24 hr

rain. Gainesville municipal water was adjusted from a pH of approximately 8.5 to 5

using 3 M HCI to mimic rainfall pH in South Florida. Rainfall was dispersed uniformly

using a TeeJetTIL HH-SS50WSQ nozzle centered under a 3 x 3 x 3 meter frame protected

from wind (Fig. 3-2). Thirty cups were distributed in a grid to test rainfall distribution

uniformity. There were no significant differences in volume within the grid. Rainfall

simulations were repeated every 3 wk, for a total of 6 rainfall simulation events.

Analysis of the source water and rainfall (pH adjusted source water) was conducted

by inductively coupled plasma atomic emission spectroscopy; total Fe and Al were

approximately 0.15 and 0.45 mg L^1 respectively. Source water and rainfall also










contained Ca (14 mg L^1) and Mg (8 mg L^)~. These Ca and Mg levels were expected

based on the municipality's well field pumping water from the Floridan aquifer which

passes through both limestone and the Hawthorne formation. In addition, the

municipality treats the water with lime. Soluble P, TP, NO3-, and NH4' were below

detectable limits (0.03, 0.03, and 0.02, and 0.06 mg L^1, respectively).





simultor







Prurface ruo
coule ctdi utr n



continue

Figure 3-2.Schmtc iga o anal iuatr(PP,20)
Rainfll bxes erepachaedfvatatmudrtesmlorta1to2sop










had~~~~~~~~ ~~~~ ben ollect.Th oeswrelf in plc ne h iultrato fo ppoimtl

half a hourwhilesubsuface low ad leahate ere cll ete. his asdneb









draining the two side and two front leachate ports, respectively. All collected samples

were weighed and then sub-sampled. A 1 L sub-sample was taken from runoff, and 250

mL sub-samples were taken from subsurface flow and leachate. Two filtered (0.45 Clm)

sub-samples (20 mL) were taken from each sub-sample, using a vacuum pump to obtain

approximately 40 mL aliquot for analysis. The sub-samples were refrigerated until P

analysis was performed, usually the following day. One 20 mL filtered sample was

acidified and refrigerated until NO3- and NH4' COuld be measured.



















Figure 3-3. Box placement under simulator.

During the 3 wk intervals between the first six simulations, the soil was watered

with pH-adjusted water to saturation and the watertable was allowed to drop to the

bottom of the boxes before watering again. A seventh simulation was run to evaluate the

effect of maintaining the soil in a saturated condition during a 4 wk period after the sixth

simulation was run. After the sixth simulation, vegetation was cut to the soil surface and

the watertable was raised to cover the surface. Water was added as needed over 4 wk to

maintain soil saturation. A plastic cover was added to reduce evaporation from the

boxes. The watertable was allowed to subside approximately 5 cm during the 2 d prior to










the 7th Simulation to facilitate moving the boxes to the simulator. The variation to the

study was made as a result of recent research at the MacArthur Agro-ecology Research

Center indicating that longer saturation times may elevate SP losses (Res. Comm., P.J.

Bohlen, MacArthur Agro-ecology Research Center).

Experimental Design

Five treatments, with four replications each were used (Fig. 3-4). Soil was placed

in the simulation boxes either as discrete layers (treatments C1, T1, and T2) or with the

two depths mixed (treatments C2 and T3). The Al-WTR was applied on the soil surface

(T1), mixed with the 0-10 cm soil depth (T2), or mixed with the combined soil depths to

a depth of 20 cm (T3).


C1* No Al-WTR applied.



T1* Al-WTR surface applied.



T2* Al-WTR incorporated into 0-10 cm soil depth.


T3** Al-WTR incorporated into 0-20 cm soil depth.



C2** No Al-WTR applied.


1 0-10 cm
10-20 cm
SAl-WVTR
S0-20 cm


*0-10 and 10-20 cm soil depths placed in box in sequence.
**0-10 and 10-20 cm soil depths mixed prior to placement in box.


Figure 3-4. Treatment descriptions









Soil Sampling and Forage Harvest

A soil core from each packed box from both depths (0-10 and 10-20 cm) was

taken prior to the start of rainfall simulations (refilled based on soil depth) and after the

sixth (5 boxes) or seventh rainfall simulations (16 boxes). Before each rainfall simulation,

the forage was harvested to a height of approximately 15 cm. The cuttings were dried,

weighed, ground to 1 mm, and stored. Upon completion of the sixth simulation, the roots

were harvested, dried, weighed, and ground from replicate 4. After the seventh

simulation, all other roots were harvested, dried, weighed, and ground. All vegetation

(initial sprigs, grass cuttings, and roots) were analyzed for total P, N, and Al.

Analytical Methods

Soil

The soil samples were dried and analyzed for WEP, TP, Mehlich-1 P, and pH.

Water extractable P in the soil was determined on air dried soil. Three grams of soil were

weighed into 50 mL centrifuge tubes and 30 mL of DDI water was added to obtain a soil

to water ratio of 1:10. The suspension was then continuously shaken on a mechanical

shaker for 1 hr and vacuum filtered (0.45 Clm). Analysis was identical to the water

samples.

Soil TP was determined by the ignition method (Anderson, 1976). One gram of

soil was weighed into a 50 mL beaker and ashed in a muffle furnace. The furnace

temperature was raised to 250oC and maintained for 30 min before ramping to 550oC for

4 hr. After ashing, the samples were brought to room temperature in a desiccator and

weighed. The cooled ash was moistened with distilled dionized water (DDI) before

adding and evaporating 20 mL of 6.0 M HCI slowly on a hotplate at approximately

120oC. Once digested, the ash was resolublized with 2.25 mL of 6.0 M HCI and









quantitatively transferred and filtered into 50 mL volumetric flasks using Whatman #41

paper. The beaker and the filter paper were each rinsed three times before bringing the

flask to volume (Anderson, 1976). Twenty milliliter sub-samples were taken and stored

in scintillation vials at room temperature. The Murphy-Riley method was used to

determine TP in solution using a Technicon TM Autoanalyzer, EPA Method 365.1 (EPA,

1993a).

Mehlich-1 extractable P, Fe, Al, Ca, and Mg were determined on air dried soil

samples from both depths (0-10 cm and 10-20 cm) of each box prior to simulations and

after simulation 7. Four grams of soil were weighed into 50 mL centrifuge tubes and 16

mL of Mehlich-1 solution (0.025 M HCI and 0.0125 M H2SO4) WAS added and shaken

continuously on a mechanical shaker for 5 min (Mehlich, 1953). The samples were then

vacuum filtered through a 0.45 Clm filter and analyzed on a TechniconThl Autoanalyzer.

The pH of the soil was measured on the supernatant of a 1:2 soil to solution ratio (DDI).

The samples were stirred and allowed to equilibrate for 30 min before determining the pH

using an Orion pH electrode.

Aluminum Water Treatment Residuals

Total P (TP) in the Al-WTR was determined via the ignition (ashing) method

(Anderson, 1976) and analyzed for P following the Murphy and Riley (1962) method.

The Al-WTR samples were digested following the EPA method 3050A and analyzed for

Al, Fe, Ca, and Mg by inductively coupled plasma atomic emission spectroscopy (ICP)

(USEPA, 1993b). Standard methods were used to determine pH (1:2 soil to solution) and

percent solids (Page, 1982).

The pH of the Al-WTR samples was measured on the supernatant 1:2 soil to

solution ratio (DDI). The samples were stirred and allowed to equilibrate for 30 min









before the pH was taken by the Orion pH electrode. Percent solids were determined for

the Al-WTR by taking the oven dried weight over the air dried weight. This is a critical

factor in determining the air dried application rate of the Al-WTR. In addition, percent

solids were determined for all water samples during the TP procedure by taking the

evaporated beaker weight over the known volume, assuming 1 mL equals 1 g.

Forage

All forage samples were digested using a modification of the standard Kj eldahl

procedure at the Forage Evaluation Support Laboratory (FESL) at the University of

Florida. Ground samples (0.25 g) were weighed into 75 mL digestion tubes and 1.5 g of

the 9:1 catalyst (K2SO4:CuSO4) WAS added. Using 4.5 mL of H2SO4 and 2 mL of H202,

the forage was digested for at least 4 hr at 3750 C (Gallaher et al., 1975). Upon

completion of the digestion, the samples were brought to volume and filtered using Fisher

screening column paper. A 20 mL sub-sample taken and was stored at room temperature

until it could be analyzed for total N (TN) by semiautomated colorimetry (Hambleton,

1977).

Runoff, Subsurface Flow, Leachate

All water samples (runoff, subsurface flow, and leachate) were vacuum filtered

through a 0.45 Clm filter within 5 hr of collection and stored in 20 mL scintillation vials at

40 C. Soluble P (SP) was analyzed using the filtered runoff samples. Total P (TP) by

digestion (see 'Soils') and pH analysis were performed on the unfiltered sample. In

addition, the runoff, subsurface flow, and leachate were analyzed for total Al using

unfiltered samples. The data were used to determine particulate P (PP) in the runoff by

subtracting soluble P (SP) from TP (TP-SP=PP). Water soluble P was determined by









EPA Method 365.1 using a Technicon TM Autoanalyzer, (USEPA, 1993a). Analyses was

performed within 48 hr of sample collection.

Water total P (TP) was determined based on a variation of the ignition (ashing)

method (Anderson, 1976). Approximately 20 to 30 mL of water was poured into a 50

mL beaker and evaporated at 100oC. The samples were then placed in a muffle furnace

and digested and sub-sampled identical to the soil samples. The Murphy-Riley method

was used to determine TP in sub-sample solution using a Technicon TM Autoanalyzer

(EPA Method 365.1, 1993a).

Total Al concentration of the water samples was determined by a nitric acid

digestion procedure (Clesceri et al., 1989). One hundred milliliters of runoff or 50 mL of

subsurface and leachate were measured into a 125 mL erlenmeyer flask with two or three

glass beads. Ten milliliters of 12 M HNO3 acid were added to the flask. The solution

was evaporated on a hot plate until approximately 10 to 20 mL remained. The flask was

then brought up to approximately 75 mL of solution using DDI water and 10 mL of 12 M

HNO3 acid was added again and allowed to evaporate to 10 to 20 mL of solution. The

solution was then filtered using Whatman #42 paper. Samples were then stored at room

temperature until analyzed for Al by atomic absorption spectroscopy.

All water samples were stored at 4oC until the pH could be measured using an

Orion pH electrode (Orion Research Inc. Boston, MA), usually within a week of

collection. The duplicate filtered water samples in 20 mL scintillation vials were acidified

using H2SO4 and stored at 40 C until analysis for ammonium and nitrate. Ammonium

was analyzed on a TechniconTM Autoanalyzer following EPA method 350.1 (USEPA,









1993c). Nitrate was analyzed on an Alpkem Corp. Rapid Flow Analyzer following

method A303-S170 (Alpkem, 1990).

Quality Assurance and Quality Control

Quality assurance and quality control (QA/QC) were met though 10% repeats,

spikes, and blanks for each procedure. Certified external standards were used for quality

control, in addition to, standard calibration curves. A 10% relative standard deviation

was required for all repeats. Less than 5% of the total samples required re-runs because

few were out-of-range values. All spike values and quality control checks fell within 90-

110% acceptance levels.

Statistical Analysis

Statistical analysis was performed using SAS version 8.2 1999-2001 (SAS Institute

Inc., Cary, North Carolina, USA). A repeated measures model was explored to estimate

and examine time (simulation) effects as well as expected correlations between

treatments and their respective control for a given simulation. Normality was checked,

and log-transformation was performed on the data (variables) to achieve normal

distribution. The General Linear Model procedure was used in order to analyze the

changes (variables) over time (simulation), between treatments, and between type. The

means values for (variables) were compared by treatment/type/simulation using the

Waller-Duncan procedure. A paired-t test was conducted in order to compare the "pre"

and "post" values for the response variables measured.














CHAPTER 4
RESULTS AND DISCUSSION

Initial Soil P Concentrations and pH

The initial soil was highly impacted with animal manure as indicated by the high

concentrations of water extractable P (WEP), Mehlich-1 extractable P, and total P (TP)

(Table 4-1). Mehlich-1 extractable P concentrations above 60 mg kg-l would require no

further addition of P from an agronomic standpoint. An un-impacted Immokalee fine

sand (A horizon) would likely have WEP concentrations of less than 1 mg kg-l and

Mehlich-1 extractable P in the range of 3 to 8 mg kg-l (Graetz and Nair, 1995). Water

extractable P concentrations and pH were similar between the 0-10 and 10-20 cm

depths; however, Mehlich-1 P and TP concentrations were greater in the 10-20 cm depth

than in the 0-10 cm depth.

Table 4-1. Initial soil mean values of P concentrations and pH of the Immokalee fine
sand used in the study.
Depth WEP Mehlich-1 P TP pH
(cm) (mg kg- ) (mg kg- ) (mg kg- )
0-10 18 879 1301 6.7
10-20 17 1034 1427 7.0

Aluminum Water Treatment Residual Characteristics

The applied Al-WTR had a moisture content of approximately 30%, a pH of 5.9,

and was dominated by Al (51,000 mg Al kg- ) (Table 4-2). Miyittah-Kporgbe (2004)

reported approximately 80% of the Al was amorphous in a similar material.

Concentrations of Fe, Ca, and Mg were low compared to Al (Table 4-2). The TP

concentration was approximately 1300 mg kg- This relatively high TP concentration is










not expected to affect the P dynamics in the soil because of its low availability. Makris

(2004) reported desorption of P by Al-WTR was less than 1%. A Hillsborough Al-WTR

as reported by O'Connor et al. (2002) had a low degree of P saturation (DPS: -0.032),

which suggests active Fe and Al for P retention.

Table 4-2. Selected characteristics of the Manatee County Al-WTR.
% Solids pH Fe Al Ca Mg TP WEP
mg kg-l mg kg-l mg kg-l mg kg-l mg kg-l mg kgl
71 5.9 1,790 51,000 580 120 1,300 BDL
SBelow detectable limit: 0.02 mg L .

Runoff, Subsurface Flow, and Leachate Characteristics

Volume

Runoff, subsurface flow, and leachate volumes were measured for each simulation

event to provide the basis for loading calculations (Table 4-3). Runoff volumes averaged

13.9 L per event with no differences in runoff volumes between treatments. Sub surface

flow volumes were considerably smaller than runoff volumes (1.2 L average) and were

uniform as well, except for treatment C2 which was greater than the other treatments.

The higher volume in C2 was attributed to slight differences in box construction and soil

packing, which resulted in slightly more standing water remaining in the C2 treatment

boxes after runoff ceased. Leachate from the bottom ports was collected only for

simulations 5 through 7. Leachate volume was smaller than subsurface flow volume but

was uniform between treatments. The uniformity of the runoff, subsurface flow, and

leachate indicate the Al-WTR does not significantly influence the soil's water holding

capacity at the rate applied.



STreatment C2 was added to the study after boxes were already constructed for all other treatments. The
boxes for treatment C2 were slightly taller than the other boxes, resulting in an increase of water held on
the soil surface after runoff ceased. This was reflected in increased subsurface flow.










Table 4-3. Average runoff, subsurface flow, and leachate volumes as influenced by
treatment from six simulations of runoff, five simulations of subsurface flow,
and two simulations of leachate.
Treatmentst
Cl T1 T2 T3 C2 Average
Collection Point ----------------------------------L---------------------
Runoff 13.97a8 13.95a 13.76a 13.88a 13.79a 13.87
Subsurface Flow 1.18b 1.13b 1.12b 1.02b 1.54a 1.20
Leachate 0.40b 0.54b 0.39b 0.63a 0.63a 0.52
tC1, no Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-WTR mixed in 0-10 cm; T3,
Al-WTR mixed in 0-20 cm; C2, no Al-WTR, mixed 0-20 cm.
8 Means within a row and followed by the same letter are not significantly different by Waller-Duncan
(p<0. 05).

pH

Runoff pH decreased from 6.76 for the first simulation to 6.02 in the sixth

simulation (Fig. 4-1A). In contrast, subsurface flow pH increased from 7.05 to 7.66

during the same period. Leachate pH, measured for the final two simulations, was slightly

greater, but similar to, subsurface flow pH. Averaged over the six simulations, there

were no differences in pH among treatments for any of the measured components

(p<0. 05; Fig. 4-1B). Results showed pH was significantly different between collection

points during each simulation and averaged over all simulations.

Phosphorus

Soluble P concentrations in surface runoff were smaller than corresponding

concentrations in subsurface flow and leachate for all treatments (Table 4-4). Soluble P

concentrations in the subsurface flow and leachate were at least 10-fold greater than in

the runoff. This can be attributed to greater volumes of rainfall appearing as runoff

(diluting the SP) than appearing as subsurface flow and leachate and less interaction with

the P-impacted soil. Runoff SP concentration was reduced by 77% by the addition of Al-

WTR to the soil surface (T1). An approximately 45% reduction in runoff SP










concentration was observed when the Al-WTR was mixed with the 0-10 cm soil depth

(T2) and 0-20 cm soil depth (T3).


(A) 8.5


-o Runoff
-m- Subsurface
--Leachate


1 2 3 4 5 6
Simulation


(B) 8.5


7.8
7-
6.5
6-
5.5 -
5-
4.5
4-


a aab~ _


a
bT


a Runoff
a Subsurface Flow
a Leachate


C1 T1 72 T3 C2
Treatment



Figure 4-1. Runoff, subsurface flow, and leachate pH values from six, five, and two
simulations, respectively. (A) Average treatment pH per. (B) Simulation
averages for each treatment. C1, no Al-WTR, not mixed; T1, surface applied
Al-WTR, not mixed; T2, Al-WTR mixed in 0-10 cm; T3, Al-WTR mixed in
0-20 cm; C2, no Al-WTR, mixed 0-20 cm. Error bars represent the standard
deviation of 4 replicates. Mean pH value within a treatment followed by the
same letter are not significantly different using the Waller-Duncan method
(p< 0. 05).










Surface application of Al-WTR (T1) did not reduce SP concentrations in

subsurface drainage or leachate (Table 4-4). Phosphorus concentrations for subsurface

flow were reduced 37% by mixing the Al-WTR in the top 10 cm of soil (T2) and 90% by

mixing the Al-WTR with the whole soil depth (T3). This difference between T2 and T3

was unexpected because the subsurface flow was collected at a depth of 10 cm and it was

anticipated that SP concentrations in the subsurface flow would be influenced primarily

by SP concentration reductions in the 0-10 cm depth. However, it appears that the SP

concentration of the subsurface flow was influenced by both the 0-10 and 10-20 cm soil

depths. This may be explained by the fact that the watertable was moving upward

through the 10-20 cm depth during part of the simulation event allowing water from the

0-10 cm depth containing Al-WTR to mix with water from the untreated 10-20 cm depth.

The SP concentration in subsurface flow and leachate in T1 increased slightly (11%) for

some unexplained reason. Treatments C1, T1, and T2 had similar leachate SP

concentrations, whereas C2 and T3 SP concentrations were significantly different. Al-

WTR treatment of the complete profile (T3) had the greatest effect on subsurface

drainage and leachate SP concentrations with reductions of 90 and 95%, respectively.

Table 4-4. Average soluble P concentrations by treatment in runoff(six events),
subsurface flow (five events), and leachate (two events).
Treatments
Cl T1 T2 T3 C2
Collection Point ------------------------------mg P L ---------------
Runoff 0.34a 0.08d 0.18c 0.16c 0.29b
Subsurface Flow 4.13b 4.57a 2.60c 0.41d 4.16b
Leachate 4.52b 4.63b 4.02b 0.36c 6.85a
t C, no Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-WTR mixed in 0-10 cm;
T3, Al-WTR mixed in 0-20 cm; C2, no Al-WTR, mixed 0-20 cm.
Means within a row and followed by the same letter are not significantly different by Waller-Duncan
(p<0. 05).









Soluble P concentrations in runoff, subsurface flow, and leachate for each

simulation event are shown in Fig. 4-2 A, B, and C, respectively. Concentrations were

greatest and most variable in the first two simulations in runoff, subsurface flow, and

leachate. These greater concentrations may be attributed to the presence of highly-labile

soil P that interacted with the rainfall during the first two simulation events. In addition,

disturbance of the soil during sampling and repacking of the simulator boxes likely

accounted for the variability observed during the first two simulations. After the first two

simulation events, concentrations of all treatments remained relatively stable for surface

runoff, subsurface flow. This consistency reinforces results from treatment averages

described previously.

Runoff SP concentrations (Figure 4-2A) were the greatest in the first two

simulations when compared to the following four simulations. In simulations 3 through

6, runoff SP concentrations in all treatments were stable and did not significantly differ

between treatments. After simulation 2, the control SP concentrations were not

significantly different and stabilized at approximately 0.3 mg L^1. When Al-WTR was

mixed with soil (T2 and T3), runoff SP concentration stabilized after two simulations at

approximately 0.15 mg L^1. When Al-WTR was surface-applied (T1), SP concentration

was less than 0.1 mg L^1 and was significantly lower than all other treatments for all

simulations.












(A).6-
0.5-
-~0.4-



0.1-
0.0
1 2 3 4 5 6
Simulation

-o-C1 -m- T1 t-AT2 T3 -m- C2


(B) 7








2 3 4 5 6
Simulation

-o C1 -m- T1 -a T2 T3 -m-C2



(C) 12

| 8 -






5 6
Simulation

-o C1 -m T1 -A T2 T3 -a- C2

Figure 4-2. Average treatment SP concentrations and standard deviations from the three
collection points. (A) Runoff: six simulations. (B) Subsurface flow: five
simulations. (C) Leachate: two simulations. C1, no Al-WTR, not mixed; T1,
surface applied Al-WTR, not mixed; T2, Al-WTR mixed in 0-10 cm; T3, Al-
WTR mixed in 0-20 cm; C2, no Al-WTR, mixed 0-20 cm.

Similar stabilization occurred with SP concentrations in subsurface flow after


simulation 2 (Figure 4-2B) for all treatments. Soluble P concentrations stabilized at


approximately 4 mg L1 after simulation 2 for the surface-applied Al-WTR (T1) and for


the controls (Cl and C2). When the Al-WTR was incorporated in to the soil (T2 and T3),










SP concentrations stabilized at 2.5 and 0.4 mg L^1, respectively. The decision to collect

leachate after simulation 4 limits interpretation of the leachate data. However, with the

exception of C2, the leachate SP concentrations were relatively consistent between

simulations 5 and 6. Soluble P concentrations for C2 were expected to be similar to C1,

but instead were higher. The reason for this is not obvious, but this difference might be

attributed to the higher water extractable P (WEP) and Mehlich-1 concentrations found in

C2 (Tables 4-11 and 4-12).

After simulation six, the soil was maintained in a flooded condition for 4 wk prior

to simulation seven. Contrary to expectations, flooding the soil reduced runoff SP

concentration by approximately 65% (Table 4-5) compared to the first six simulations

(Table 4-4). This effect was believed to be a result of longer interaction time with

soil/Al-WTR sorption sites. Starting the rainfall event with the soil flooded likely

minimized the rainfall interaction with the soil solution which resulted in lower P

concentrations in the runoff. Flooding of the soil also slightly reduced subsurface SP

concentrations compared to the previous six simulations. No impact of flooding on SP

concentrations was observed for the leachate.

Table 4-5. Average soluble P concentrations in runoff, subsurface flow, and leachate
after 4 wk of flooded conditions (Simulation 7).
Treatments
Cl T1 T2 T3 C2
Collection Point ------------------------------mg P L^1-----------------------------
Runoff 0.13a8 0.04b 0.04b 0.03b 0.15a
Sub surface 3.53a 3.43a 1.61b 0.20c 2.69a
Leachate 3.96b 4.20b 3.49bc 0.23c 7.95a
t C, no Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-WTR mixed in 0-10 cm;
T3, Al-WTR mixed in 0-20 cm; C2, no Al-WTR, mixed 0-20 cm.
Means within a row and followed by the same letter are not significantly different by Waller-Duncan
(p<0. 05).










Runoff, subsurface flow, and leachate volumes for simulation 7 followed the same

trend of consistency between treatments as was observed for the first 6 simulations

(Table 4-6A). Flooding the soil lowered the average runoff pH from 6.8 (simulations 1-

6) to 6.3 (simulation 7) (Table 4-6B). Average subsurface flow and leachate pH values

decreased from approximately 7.6 to 7.3.

Table 4-6. Runoff, subsurface flow, and leachate collected during simulation 7. (A)
Volume. (B) pH.
(A) Treatments
Cl T1 T2 T3 C2
Collection Point -----------------------------------L--------------------
Runoff 15.81b~ 16.78b 15.23c 15.83b 16.63b
Sub surface 0.77a 0.86a 1.07b 1.11ab 1 .22b
Leachate 0.42a 0.73a 0.36a 0.63a 0.57a

(B) Treatments
Cl T1 T2 T3 C2
Collection Point ---------------------------------pH--------------------
Runoff 6.34a~ 6.10a 6.44a 6.30a 6.38a
Sub surface 7.37b 7.42b 7.39b 7.31b 7.19b
Leachate 7.44c 7.44c 7.59c 7.35c 7.34c
tCl, no Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-WTR mixed in 0-10 cm; T3, Al-
WTR mixed in 0-20 cm; C2, no Al-WTR, mixed 0-20 cm.
~Means within a row and followed by the same letter are not significantly different by Waller-Duncan (p<0. 05).

Soluble P loss per box from each treatment per rainfall event was calculated by

multiplying volume by concentration (Table 4-7). The same trend was observed as with

SP concentrations, which was expected based on volume similarities among treatments

(Table 4-6A). These results emphasize the importance of treating P-impacted soils to

reduce risk ofP loss. Mixing 2.5% Al-WTR with the P-impacted soil can reduce field

loss of SP by approximately 80% compared to 30% by surface applied Al-WTR. Surface

application and partial mixing of the impacted zone (T 1 and T2, respectively) had less

impact compared to total mixing of Al-WTR with the P-impacted soil (T3). The

relatively lower volumes of subsurface and leachate pose an equal risk of SP loss to that









of runoff because what is lacking in volume is compensated by concentration. A rough

estimate of SP loss on a hectare basis can be calculated, assuming that similar P losses

would be encountered over larger areas (Table 4-7). The expected loss per rainfall event

in an untreated field would be 3 50 to 500 g ha- Partial treatment of the impacted zone

could reduce P loss approximately 200 to 300 g ha l. Soluble P loss could be reduced to

less than 100 g ha-l if Al-WTR was incorporated throughout the P-impacted layer.

Soil mixing in C2 did not reduce P loss as was found by Miyittah (2004), because

both the 0-10 and 10-20 cm depths were highly P-impacted. Miyittah (2004) mixed a P-

impacted soil with an un-impacted E horizon, which in effect, diluted the P-impacted soil.

Table 4-7. Soluble P mass loss (SP concentration volume) averaged over six rainfall
events and estimated SP loss per hectare per rainfall event.
Treatments
Collection Point Cl T1 T2 T3 C2
Runoff (mg P box- ) 4.3 6a" 1.07d 2.22c 1.96c 3.68b
Subsurface (mg P box- ) 4.75b 4.92b 2.76c 0.45d 6.07a
Leachate (mg P box- ) 1.69b 2.37b 1.50b 0.20c 4.61a
Total SP loss (mg P box- ) 10.80 8.36 6.48 2.61 14.36
SP loss (g P ha- ) 360 279 216 87 479
t C, no Al-WTR, not mixed: T1, surface applied Al-WTR, not mixed: T2, Al-WTR mixed in 0-10 cm;
T3, Al-WTR mixed in 0-20 cm; C2, no Al-WTR, mixed 0-20 cm.
Means within a row and followed by the same letter are not significantly different by Waller-Duncan
(p 0.005).

Particulate P (PP) concentration was calculated in runoff, subsurface flow, and

leachate by subtracting SP from TP (Table 4-8). Runoff PP concentrations ranged from

0. 1 to 0.3 mg L 1. On average, PP accounted for approximately 20 and 60% of TP in the

controls and the Al-WTR treatments, respectively. Therefore, PP should also be a

concern on minimally sloped landscapes, particularly if Al-WTR has been applied.

Particulate P concentrations in the subsurface flow and leachate were below detectable

limits. Although PP concentrations in runoff showed marked trends among treatments,

the differences were not significant because of the high standard deviations among










treatments. This is believed to be a result of small variations in soil surface levels

compared to the box edge rather than treatment effects. This would affect the amount of

particulate matter leaving, or conversely, retained, at the box edge where runoff was

collected.

Table 4-8. Total, soluble, and particulate P concentrations in runoff for each treatment
averaged over six rainfall events.
Treatments
Cl T1 T2 T3 C2
--------------------------------mg P L^----------------
Total P 0.47a~ 0.21b 0.38ab 0.47a 0.34ab
Soluble P 0.34a 0.08d 0.18c 0.16c 0.29b
Particulate P 0.13a 0.13a 0.20a 0.31a 0.05a
t C, no Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-WTR mixed in 0-10 cm;
T3, Al-WTR mixed in 0-20 cm; C2, no Al-WTR, mixed 0-20 cm.
Means within a row and followed by the same letter are not significantly different by Waller-Duncan
(p<0. 05).

Nitrate- and Ammonium-Nitrogen

Ammonium-N concentrations in runoff, subsurface flow, and leachate averaged

less than 1 mg L^1 and there were no differences among treatments (Table 4-9A). Nitrate-

N concentrations (Table 4-9B) were also below 1 mg L^1 and showed no differences

between treatments with two exceptions. The exceptions were greater NO3-N

concentrations in subsurface flow and leachate in T1 compared to all other treatments.

There are no obvious explanations for these differences at this time. Under typical

production conditions, N fertilizer would be added to maintain production and forage

nutritive value. However, to simplify this experiment, no additional fertilization was

used.










Table 4-9. Average N concentrations in runoff, subsurface flow, and leachate from six,
five, and two simulations, respectively. (A) NH4-N. (B) NO3-N.
(A) Treatmentst
Clt T1 T2 T3 C2
Collection Point --------------------------mg NH4-N L 1-------------------------
Runoff 0. 19a~ 0.19a 0.18a 0.23a 0.19a
Sub surface 0.38a 0.38a 0.35a 0.30a 0.33a
Leachate 0.35a 0.36a 0.42a 0.47a 0.44a

(B) Treatments
Clt T1 T2 T3 C2
Collection Point --------------------------mg NO3-N L -------------
Runoff 0.15a~ 0.14a 0.14a 0.13a 0.12a
Sub surface 0.23b 0.77a 0.10b 0.14b 0.13b
Leachate 0.12b 0.74a 0.11lb 0.10b 0.08b
t C, no Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-WTR mixed in 0-10 cm;
T3, Al-WTR mixed in 0-20 cm; C2, no Al-WTR, mixed 0-20 cm.
~Means within a row and followed by the same letter are not significantly different by Waller-Duncan
(p<0. 05).

Total Calcium, Magnesium, Iron, and Aluminum Concentrations

Calcium and Mg concentrations were in the order of runoff < subsurface flow <

leachate (Table 4-10). There were no significant differences in Ca and Mg

concentrations in runoff among treatments. Calcium and Mg concentrations were

different among treatments for subsurface flow and leachate. Aluminum-WTR

treatments had higher concentrations of Ca in subsurface flow and leachate which is

likely due to the Ca contained in the Al-WTR (580 mg L^)~. Iron concentrations were

similar in runoff, subsurface flow, and leachate, and did not vary between treatments.

Total Al concentrations were highly variable and showed no meaningful differences

among treatments. However, a trend of higher Al concentrations was observed in

leachate from treatments where Al-WTR was incorporated into the soil (T2 and T3).

Random samples were tested for soluble Al from all collection points, treatments, and

simulations and all concentrations were below detection limits (0.3 mg L^)~. This

suggests that essentially all of the Al was in particulate form.










Table 4-10. Average total Ca, Mg, Fe, and Al concentrations in runoff, subsurface flow,
and leachate from six, five, and two simulations, respectively. (A) Runoff.
(B) Subsurface flow. (C) Leachate.
(A) Runoff
Ca Mg Fe Al
Treatm entst-------------- g L -------- -----
Cl 15.8 8.83 0.27 0.66
T1 16.3 8.90 0.53 1.06
T2 16.4 8.44 0.43 1.62
T3 17.6 8.68 0.51 2.68
C2 15.7 8.36 0.23 2.71


(B) Subsurface Flow
Ca Mg Fe Al
Treatments -------------------------mg L^'-------------------------
Cl 64.3c' 22.1b 0.23 0.00b
T1 80.2b 26.8ab 0.42 1.22a
T2 91.1ab 30.6a 0.34 0.00b
T3 95.9a 31.6a 0.38 0.00b
C2 49.0d 16.4c 0.54 0.76a


(C) Leachate
Ca Mg Fe Al
Treatments -------------------------mg L^ -----------------------
Cl 95.2b' 45.8b 0.19 0.00a
T1 94.4b 48.7b 0.09 0.00a
T2 138.8ab 67.9a 0.15 1.94b
T3 199.3a 63.0a 1.14 5.67b
C2 87.7b 28.1c 0.12 3.70b
t C, no Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-WTR mixed in 0-10 cm;
T3, Al-WTR mixed in 0-20 cm; C2, no Al-WTR, mixed 0-20 cm.
Means within a column and followed by the same letter are not significantly different by Waller-Duncan
(p<0. 05).

Pre- and Post-simulation Comparisons

Soil Phosphorus

When Al-WTR was incorporated into the soil, WEP decreased by approximately

70% i.e., T2 (0-10 cm depth) and both depths in T3, compared to the respective controls

(Table 4-11). There was no change in soil WEP concentrations in any of the treatments


prior to or after the study (paired T Test; p< 0.05).










Table 4-11. Pre- and post-simulation water extractable P concentrations in the 0-10 and
10-20 cm soil depths.
0-10 cm depth 10-20 cm det
Treatmentst Pre-simulation~ Post-simulation Pre-simulation Post-simulation

Cl 21 a 20 a 24 a 26 a
T1 14 c 14 b 24 a 23 ab
T2 7 d 7 c 19 b 18 b
T3 6d 7c 7c 6c
C2 25 a 24 a 24 a 25 a
t C1, no Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-WTR mixed in 0-10 cm;
T3, Al-WTR mixed in 0-20 cm; C2, no Al-WTR, mixed 0-20 cm.
"Means within a column followed by the same letter are not significantly different by Waller-Duncan
(p<0. 05).

Application of Al-WTR had essentially no effect on Mehlich-1 extractable P (Table

4-12). Apparently, Mehlich-1 extractant removes P adsorbed to the soil as well as that

adsorbed to Al-WTR. There were few differences in pre- and post-simulation

concentrations statistically compared using the paired T Test (p<0. 05).

Table 4-12. Pre- and post-simulation Mehlich-1 P concentrations in the 0-10 and 10-20
cm soil depths.
0-10 cm depth 10-20 cm depth
TretmetstPre-simulation'' Post-simulation Pre-simulation Post-simulation

Cl 679 AB 785 B 882 A 882 BC
T1 635 AB 774 B 825 AB 1017 A
T2 614 B 709 B 760b AB 954a AB
T3 779 A 770 B 720 B 795 C
C2 777a A 958b A 823 AB 903 B
t C1, no Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-WTR mixed in 0-10 cm;
T3, Al-WTR mixed in 0-20 cm; C2, no Al-WTR, mixed 0-20 cm.
Means for each soil depth within a row followed by the same lower case letter are not significantly
different by Waller-Duncan (p< 0. 05).
SMeans for each soil depth within a column followed by the same upper case letter are not significantly
different by Waller-Duncan (p< 0. 05).

Soil pH

Soil pH decreased slightly by the end of the study (Table 4-13). However,

application of Al-WTR had essentially no effect on soil pH before or after simulation










events. It appears the soil was well-buffered due to the Ca and Mg accumulated from the

manure applications.

Table 4-13. Pre- and post-simulation soil pH values in the 0-10 and 10-20 cm soil
depths.
0-10 cm depth 10-20 cm depth
Treatments Pre-simulation Post-simulation Pre-simulation Post-simulation
Cl 6.9a A 6.6b A 7.0a AB 6.7b A
T1 6.7 AB 6.6 A 7.0a A 6.6b A
T2 6.9 B 6.6 A 6.9a ABC 6.6b A
T3 7.0a A 6.6b A 6.8 C 6.6 A
C2 6.9a A 6.6b A 6.8a BC 6.6b A
t C, no Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-WTR mixed in 0-10 cm;
T3, Al-WTR mixed in 0-20 cm; C2, no Al-WTR, mixed 0-20 cm.
Means for each soil depth within a row followed by the same lower case letter are not significantly
different by Waller-Duncan (p< 0. 05).
SMeans for each soil depth within a column followed by the same upper case letter are not significantly
different by Waller-Duncan (p< 0. 05).

Forage

Shoot yields were smaller for the surface-applied Al-WTR (T1) than for

incorporated Al-WTR (T2 and T3) treatments for the initial three harvests (Table 4-14).

This might suggest that surface-applied Al-WTR was detrimental to stargrass growth.

However, this same trend occurred for control C2. This suggests that the differences may

be due to initial stand establishment rather than toxicity effects. After the first two

simulations (63 d of growth), the stargrass became well-established. For simulations 4-6,

shoot yields were similar between treatments. There was no visual evidence of adverse

effects on stargrass shoot growth. In addition, root mass of all treatments was

approximately 30 g box-l during the experiment (Figure 4-3). Root biomass was not

significantly different among treatments; therefore Al-WTR does not appear to adversely

affect stargrass roots.










Table 4-14. Stargrass shoot biomass harvested prior to each simulation.
Treatmentst Sim 1 Sim 2 Sim 3 Sim 4 Sim 5 Sim 6' Avg
-------------------------------------g box-
Cl 35a" 17a 38a 16a 8a 57a 23
T1 17b 8b 24b 19a 4b 41a 14
T2 35a 21a 35ab 18a 7a 50a 23
T3 31a 14a 30ab 17a 7a 56a 20
C2 14b 6b 28ab 21a 9a 47a 15
SC1, no Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-WTR mixed in 0-10 cm;
T3, Al-WTR mixed in 0-20 cm; C2, no Al-WTR, mixed 0-20 cm.
Simulation 6 data includes all above-ground biomass. Simulation 1-5 data are shoot biomass above 15 cm
height.
"Shoot biomass was compared (column) between treatments using a log transformation followed by
Waller-Duncan method. Treatments with the same letter are not significantly different (p<0. 05).


140 -a b a ab b
120-
S100-
a 80 aShoots
E 60- a a a a a gRos Roots
(3 40-



C1 T1 T2 T3 C2
Treatment



Figure 4-3. Total shoot and root growth per treatment averaged over six simulations. C1,
no Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-
WTR mixed in 0-10 cm; T3, Al-WTR mixed in 0-20 cm; C2, no Al-WTR,
mixed 0-20 cm. Error bars represent the standard deviation of 4 replicates.
Treatments means within a plant fraction and followed by the same letter are
not significantly different using the Waller-Duncan method (p<0. 05).

Shoot and root P concentrations among all treatments averaged over the six

simulations were not different (p<0. 05; Figure 4-4A). During the first two simulations,

the stargrass contained approximately 3 mg of P g-l of forage and declined by

approximately 0.5 mg uniformly across all treatments by the end of the experiment

(Figure 4-4B). The original planting material contained approximately 3.2 mg g- The

decline was expected based on seasonal change in growth (August to January).







48


Simulation 1 began in August and simulation 6 ended in December. Root P

concentrations were approximately 1.7 mg g-l and did not significantly vary among

treatments (Figure 4-4A).


(A) 4-









C1 T1 T2 T3 C2
Treatments



(B) 4









1 2 3 4 5 6
Ha rvest

Figure 4-4. Average stargrass TP concentrations per treatment averaged over all
simulations (A) and per simulation averaged over all treatments (B). C1, no
Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-WTR
mixed in 0-10 cm; T3, Al-WTR mixed in 0-20 cm; C2, no Al-WTR, mixed
0-20 cm. Error bars represent the standard deviation of 4 replicates.

Shoot and root TN concentrations also were not affected by treatment (p <0. 05;

Figure 4-5A). Shoot TN concentration was approximately 25 mg N g- in the planting

material and for the first two simulations (Figure 4-5B). However, with no additional N

input, the N concentration declined to approximately 10 mg N g- Root TN

concentrations at the end of the study were similar for all treatments (Figure 4-5A).







49



(A) 30-

25-









20-

E 15-





C1 21 32 4 56


Figue 45. verae sargass N cncetrain e treatment aeae oe l

siultin (A n e iuain vrgdoe l retet B.Cn
AlWR o ie;Tsraeaple lWR o ie;TA-T











tyial eandblw4 gg Figure 4-6A, B).ag Theseas concentrations we ramn vrgdoere atth

low endof thetyical rA)ange (601450 ppm) ftor grasses (Pendiasl andPendis B C, 201)

Root Al WR cocnrtons ranged fo 7 to, 310ac apge g1- R oft forge Tese value alsofal

within the tpcal rne for forsrae plants.Bcueo the hg standard deviation, therepicte








were no significant differences among treatments (p<0. 05). Contamination is believed to

be a factor even though roots were washed multiple times to remove soil and Al-WTR







50


particles. No stunted roots were observed. Aluminum concentration (Figure 4-6B)

exhibited a decline that was similar to P and N concentrations in simulations 3 through 5.

The increase in concentration in simulation 6 is attributed to shoot contamination when

cutting shoots close to the soil surface; which is supported by the large standard

deviation.


(A)1600
1400
1200
S1000

800
6 00
400
200
0





(B) 250-

200-



150-



0-


I Shoots
I Roots


C1 T1 T2 T3 C2
Treatments


~i~-I~J


1 2 3
Harvest


4 5


Figure 4-6. Average stargrass Al concentrations per treatment averaged over all
simulations (A) and per simulation averaged over all treatments (B). C1, no
Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-WTR
mixed in 0-10 cm; T3, Al-WTR mixed in 0-20 cm; C2, no Al-WTR, mixed
0-20 cm. Error bars represent the standard deviation of 4 replicates.
Treatments means within the root fraction and followed by the same letter are
not significantly different using the Waller-Duncan method (p<0. 05).















CHAPTER 5
CONCLUSIONS

Previous research has shown that land applying Al-WTR is effective in reducing

soluble P loss from agricultural fields. This study was developed to provide Al-WTR

management information on manure-impacted sandy soils with a fluctuating high

watertable. The research evaluated different methods of applying Al-WTR (surface and

incorporated) as well as different incorporation depths, P loss over time, and potential

effects on forage productivity and chemical composition. Runoff boxes were filled with

a P-impacted sandy soil and the respective Al-WTR application treatments. The rate of

Al-WTR application (2.5% of dry weight of soil) was selected based on Miyittah's

(2004) recommendation for a practical field application rate that would significantly

reduce soluble P concentrations in the soil solution. Boxes were placed under a rainfall

simulator to create uniform rainfall distribution at known intervals and duration.

Phosphorus loss was examined from the three collection points (runoff, subsurface flow,

and leachate) because of potential water movement above a restricting layer (spodic).

Concerns about P deficiency and Al toxicity in stargrass were addressed throughout the

duration of the study.

The application of Al-WTR, whether surface applied or mixed within the profile,

impacted SP loss. Surface applied Al-WTR was more effective in reducing SP in runoff

than when incorporated. However, incorporated Al-WTR was more effective in reducing

SP in subsurface flow and leachate than was surface Al-WTR application. Care must be

taken to ensure complete incorporation of Al-WTR throughout the P-impacted layer, as









Al-WTR is only effective in reducing SP concentrations when it is in contact with the

impacted soil. To achieve the best results for reducing P loss in both surface runoff and

subsurface flow/leachate from highly impacted soils, Al-WTR should be first mixed with

the impacted soil depth to reduce subsurface flow/leachate P loss AND then added to the

soil surface to minimize P loss in runoff. For an un-impacted area with low initial soil P

concentration intended for manure application, surface application of Al-WTR would

likely suffice to minimize P loss.

Application of Al-WTR at 2.5% of soil weight did not adversely affect forage yield

or quality of stargrass, based on the uniform values of yield and P, N, and Al

concentrations between treatments. Field scale forage studies are needed to validate box-

scale results.

Effective P control on intensively managed agricultural land is imperative. The use

of Al-WTRs is not a single source solution. Rather, it is an effective management tool in

an intricate comprehensive management plan. This study attempted to mimic field

conditions to provide realistic results. However, short- and long -term Hield studies are

needed before Al-WTR application can be included in best management practices.
















LIST OF REFERENCES


Alpkem Corp. 1990. Rapid Flow Analyzer Methodology. Method A303-S170.
Clackamas, OR.

Anderson, D.L., O.H. Tuovinen, A. Faber, and I. Ostrokowski. 1995. Use of soil
amendments to reduce soluble phosphorus in dairy soils. Ecol. Eng. 5: 229-246.

Atalay, A. 2001. Variation in phosphorus sorption with soil particle size. Soil and Sed.
Contam. 10: 317-335.

Basta, N.T., and J.J. Sloan. 1999. Bioavailability of heavy metals in strongly acidic soils
treated with exceptional quality biosolids. J. Environ. Qual. 28: 633-63 8.

Ballard, R., and J.G.A. Fiskell. 1974. Phosphorus retention in coastal-plain forest soils:
Relationship to soil properties. Soil Sci. Soc. of Amer. J. 38: 250-255.

Boggess, C.F., E.G. Flaig, and R.C. Fluck. 1995. Phosphorus budgets for Lake
Okeechobee tributary watersheds. Ecol. Eng. 5: 143-162.

Breeuwsma, A., and S. Silva. 1992. Phosphorus fertilization and environmental effects
in the Netherlands and the Po regions (Italy). Rep. 57. Agric. Res. Dep., The
Winand Staring Centre for Integrated Land, Soil, and Water Res., Wageningen, the
Netherlands.

Brown, E., and J.B. Sartain. 2000. Phosphorus retention in United States Golf
Association (USGA) greens. Soil Crop Sci. Soc. Florida Proc. 59: 112-117.

Bugsbee, G.J., and C.R. Frink. 1985. Alum sludge as a soil amendment: Effects on soil
properties and plant growth. New Haven Connecticut Agric. Exp. Stn. Bull. 827.

Buss, G.R., J.A. Lutz, Jr., and G.W. Hawkins. 1975. Yield response of alfalfa cultivars
and clones to several pH levels in Tatum subsoil. Agron. J. 67: 331-334.

Clesceri, L.S., A.E. Greenberg, and R.R. Trussell (ed.). 1989. 3030D, Preliminary
digestion for metals. In: Standard Methods for the Examination of Water and
Wastewater: 17th ed. Amer. Public Health Assoc. Washington, D.C.

Cornell, D.A., and G.P. Westerhoff. 1981. Management of water treatment plant
sludges. In: Sludge and its Ultimate Disposal. (Ed. J.A. Borchardt). Ann Arbor
Sci. Publishers. Ann Arbor, MI.










Dayton, E.A., and N.T. Basta. 2001. Characterization of drinking water treatment
residuals for use as a soil substitute. Water Environment Research 73: 52-57.

Delhaize, E., and P.R. Ryan. 1995. Aluminum toxicity and tolerance in plants. Plant
Physiol. 107: 315-321.

Elliott, H.A., and B.A. Dempsey. 1991. Agronomic effects of land application of water-
treatment sludges. J. Amer. Water Works Assoc. 83: 126-131.

Elliott, H.A., G.A. O'Connor, P. Lu, and S. Brinton. 2002. Influence of water treatment
residuals on phosphorus solubility and leaching. J. Environ. Qual. 31: 1362-1369.

Elliott, H.A., B.A. Dempsey, D.W. Hamilton, and J.R. Dewolfe. 1990. Land application
of water treatment sludges; Impacts and management. Amer. Water Works Assoc.
Res. Denver, CO.

Esteves, M., O. Planchon, J.M. Lapetite, N. Silvera, and P. Cadet. 2000. The 'emire'
large rainfall simulator: design and field testing. Earth Surf. Processes Landforms
25: 681-690.

Farmer, EE. 1973. Relative detachability of soil particles by simulated rainfall. Soil Sci.
Soc. of Amer Proc. 37: 629-633.

Faulkner, S.P., and C.J. Richardson. 1989. Physical and chemical characteristics of
freshwater wetland soils. pp. 41-72. In: Hammer, D.A. (ed.). Constructed
Wetlands for Wastewater Treatment. Lewis Publishers. Chelsea, MI.

Foy, C.D. 1984. Physiological effects of hydrogen, aluminum, and manganese toxicities
in acid soils. In: Soil acidity and liming, (Ed. F. Adams). pp. 57-97. Agron.
Monograph. 12, 2nd ed. Amer. Soc. Agron., Crop. Sci. Soc. Amer., and Soil Soc.
Amer., Madison, WI.

Foy, C.D., R.L. Chaney, and M.C. White. 1978. Physiology of metal toxicity in plants.
Ann. Rev. Plant Physiol. 29: 511-566.

Foy, C.D., A.L. Fleming, and G.C. Gerloff. 1972. Differential aluminum tolerance in
two snapbean varieties. Agron. J. 64: 815-818.

Flaig, E.G., and K.R. Reddy. 1995. Fate of phosphorus in the Lake Okeechobee
watershed, FL, USA: overview and recommendations. Ecol. Eng. 5: 127-142.

Flaig, E.G., and K.E. Havens. 1995. Historical trends in the Lake Okeechobee
ecosystem. I. Land use and nutrient loading. Arch. Hydrobiol. Suppl. 107: 1-24.

Florida Agriculture Statistical Directory. 2004. Florida Department of Agricultural and
Consumer Services. Tallahassee, FL.










Fluck, R.C., C. Fonyo, and E. Flaig. 1992. Land-use-based phosphorus balances for
Lake Okeechobee, Florida, drainage watersheds. Appl. Eng. Agric. 8: 813-820.

Froelich, P.N. 1988. Kinetic control of dissolved phosphate in natural rivers and
estuaries: A primer on the phosphate buffer mechanism. Limnol. Oceanogr. 33:
649-668.

Frossard, E., M. Brossard, M.J. Hedley, and A.K. Metherell. 1995. Reactions
controlling the cycling of P in soils. In: Phosphorus in the Global Environment
Transfers, Cycles, and Management. (Ed. H. Tiessen). John Wiley & Sons, New
York, NY.

Gale, P.M., K.R. Reddy, and D.A. Graetz. 1994. Phosphorus retention by wetland soils
used for treated wastewater disposal. J. Environ. Qual. 23: 3 70-3 77.

Gallaher, R.N., C.O. Weldon, and J.G. Futral. 1975. An aluminum block digester for
plant and soil analysis. Soil Sci. Soc. Amer. Proc. 39: 803-806.

Gallimore, L.E., N.T. Basta, D.E. Storm, M.E. Payton, R. H. Huhnke, and M.D. Smolen.
1999. Water treatment residual to reduce nutrients in surface runoff from
agricultural land. J. Environ. Qual. 28: 1474-1478.

Gatewood, S.E., and P.B. Bedient. 1975. Drainage density in the Lake Okeechobee
drainage area. Report, Division of State Planning, Florida Department of
Administration. Tallahassee, FL.

Graetz, D.A., and V.D. Nair. 1995. Fate of phosphorus in Florida Spodosols
contaminated with cattle manure. Ecol. Eng. 5: 163-182.

Hambleton, L.G. 1977. Semiautomated method for simultaneous determination of
phosphorus, calium, and crude protein in animal feeds. J.A.O.A.C. 60: 845-852.

Haustein, G.K., T.C. Daniel, D.M. Miller, P.A. Moore, Jr., and R.W. McNew. 2000.
Aluminum-containing residuals influence high-phosphorus soils and runoff water
quality. J. Environ. Qual. 29: 1954-1959.

Haug, A. 1984. Molecular aspects of aluminum toxicity. CRC Crit. Rev. Plant Sci. 1:
345-373.

He, Z.L., A.K. Alva, Y.C. Li, D.V. Calvert, and D.J. Banks. 1999. Sorption-desorption
and solution concentration of phosphorus in a fertilized sandy soil. J. Environ.
Qual. 28: 1804-1810.

Hoddinott, J., and C. Richter. 1987. The influence of aluminum on photosynthesis and
translocation in french bean. J. Plant Nutr. 10: 443-454.










Hodges, J.R., G.K. Kirk, R.L. Shirley, F.M. Peacock, J.F. Easley, H.L. Breland, and F.G.
Martin, 1967. Phosphorus fertilization of pangola grass pastures, direct and
residual effects. Research Report, Range Experiment Station, Univ. of FL, Ona,
FL.

Holford, I.C.R., C. Hird, and R. Lawie. 1997. Effects of animal effluents on the
phosphorus sorption characteristics of soils. Aust. J. Soil Res. 35: 365-373.

Howard, F. 2006. Pulling out P. Hay and Forage. March 2006: 52.

Humphry, J.B., T.C. Daniel, D.R. Edwards, and A.N. Sharpley. 2002. A portable
rainfall simulator for plot-scale runoff studies. Appl. Eng. Agric. 18: 199-204.

Ippolito, J.A., K.A. Barbarick, D.M. Heil, J.P. Chandler, and E.F. Redente. 2003.
Phosphorus retention mechanisms of a water treatment residual. J. Environ. Qual.
32: 1857-1864.

Ippolito, J.A., K.A. Barbarick, and E.F. Redente. 2002. Combinations of water treatment
residuals and biosolids affect two range grasses. Commun. Soil Sci. Plant Anal.
33: 831-844.

Jarvis, S.C., and D.J. Hatch. 1986. The effects of low concentrations of aluminum on the
growth and uptake of nitrate-N by white clover. Plant Soil 95: 43-55.

Kinraide, T.B. 1991. Identity of the rhizotoxic aluminium species. Plant Soil. 134:
167-178.

Kleinman, P.J.A., and A.N. Sharpley. 2003. Effect of broadcast manure on runoff
phosphorus concentration over successive rainfall events. J. Environ. Qual. 32:
1072-1081.

Kleninman, P.J.A., A.N. Sharpley, B.G. Moyer, and G.F. Elwinger. 2002. Effect of
mineral and manure phosphorus sources on runoff phosphorus. J. Environ. Qual.
31: 2026-2033.

Klimashe, E.L. 1970. On role of roots as determining different tolerance of genetically
related plant forms toward Al3+. Agrochimica 14: 232-241.

Lindsay, W.L., P.L.G. Vlek, and S.H. Chien. 1989. Phosphate minerals. pp. 1089-
1130. In: Minerals in Soils Environments. (Ed. J.B. Dixon, and S.B. Weed). Soil
Sci. Soc. Amer. J., Madison, WI.

McCaffery, P.M., W.M. Hinkley, R. MacGill, and G.D. Cherr. 1976. Report of
investigations in the Kissimmee-Lake Okeechobee Watershed. FL Dept. Environ.
Reg. Tech. Series, Vol. 2, No. 2, Tallahassee, FL.

Mehlich, A. 1953. Determination of P, Ca, Mg, K, Na, and NH4. Soil Testing Div.
Publ. 1-53. North Carolina Department of Agriculture, Raleigh.










Meyer, L.D.,and D.L. McCune. 1958. Rainfall simulator for runoffplots. Agric. Eng.
39: 644-648.

Miller, W.P. 1987. A solenoid-operated, variable intensity rainfall simulator. Soil Sci.
Soc. of Amer. J. 51 : 832-834.

Miner, C.L. 2001. Storage and partitioning of soil phosphorus in the Orlando Easterly
Wetland treatment system. M. Sci. Thesis, Univ. of FL, Gainesville, FL. 32611-
0510.

Mitsch, W.J., and J.G. Gosselink. 1993. Wetlands, 2nd ed. Van Nostrand Reinhold
Company, NY.

Miyittah-Kporgbe, M. 2004. Phosphorus immobilization in manure-impacted soil with
aluminum-based drinking water treatment residual. M.Sci. Thesis, Univ. of FL,
Gainesville, FL. 32611-0510.

Moore, P.A., Jr., P.B. DeLaune, D.E. Carman, T.C. Daniel, and A.N. Sharpley. 2000.
Development of phosphorus index for pastures. pp. 158-165. In: Proc. 2000 Natl.
Poultry Waste Management Symp. (Ed. J.P. Blake, and P.H. Patterson). Auburn
Press, Auburn, AL.

Moore, P.A. Jr., D.M. Miller, T.C. Daniel, B.R. Shreve, and D.R. Edwards. 1994.
Decreasing phosphorus solubility and inhibiting ammonia volatilization in poultry
litter with alum. Poultry Science 73: 151.

Munn, J.R., and G.L. Huntington. 1976. A portable rainfall simulator for erodibility and
infiltration measurements on rugged terrain. Soil Sci. Soc. ofAmer. J. 40: 622-
624.

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

Nair, V.D., and D.A. Graetz. 2002. Phosphosrus saturation in spodosols impacted by
manure. J. Environ. Qual. 31: 1279-1285.

Nair, V.D., D.A. Graetz, and K.M. Portier. 1995. Forms of phosphorus in soil profiles
from dairies of south Florida. Soil Sci. Amer. J. 59: 1244-1249.

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.

National Phosphorus Research Project (NPRP). 2001. Rain simulator protocols.
(Accessed 11/3/06).
http://www. seral17. ext.vt. edu/Documents/Nati onal P_protocol .pdf.










Neff, E.L. 1979. Why rainfall simulation? In: Proceedings of Rainfall Simulator
Workshop. Tucson, AZ. USDA-SEA ARM-W-10: 3-7.

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

O'Connor, G.A., H.A. Elliott, and P. Lu. 2002. Characterizing water treatment residuals
phosphorus retention. Soil and Crop Sci. Soc. FL Proc. 61: 67-73.

Ohki, K. 1986. Photosynthesis, chlorophyll, and transpiration responses in aluminum
stressed wheat and sorghum. Crop Sci. 26: 572-575.

Olila, O.G., and K.R. Reddy. 1995. Influence of pH on phosphorus retention in oxidized
lake sediments. Soil Sci. Soc. of Amer. J. 59: 946-959.

Osborne, G.J., J.E. Pratley, and W.P. Stewart. 1981. The tolerance of subterranean
clover (Trifolium-Subterraneum L) to aluminum and manganese. Field Crops Res.
3: 347-358.

Page, A.L. 1982. Methods of Soil Analysis. Part 2. Chemical and Microbiological
Properties. 2nd ed. Agron. Monogr. 9. ASA and SSSA. Madison, WI.

Reddy, K.R., E.G. Flaig, and D.A. Graetz. 1996. Phosphorus storage capacity of
uplands, wetlands and streams of the Lake Okeechobee watershed, Florida. Agric.
Ecosystems & Environ. 59: 203-216.

Reddy, K.R., and W.H. Smith (ed). 1987. Aquatic plants for water treatment and
resource recovery. Magnolia Publishing, Orlando, FL.

Riley, S.J., and Hancock F. 1997. A rainfall simulator for hydrologic and erosion
experiments on mines, with an example of its applications at Ranger Uranium
Mine, Northern Territory, Australia. The Australasian Institute of Mining and
Metallurgy Proceedings. 1: 3-8.

Romkens, M.J.M., L.F. Glenn, D.W. Nelson, and C.B. Roth. 1975. A laboratory rainfall
simulator for infiltration and soil detachment studies. Soil Sci. Soc. of Amer. Proc.
39: 158-160.

SAS Institute, 2001. The SAS System for Windows. Version 8 Release 8.2 SAS Inst.
Cary, NC.

Samac, D.A., and M. Tesfaye. 2003. Plant improvement for tolerance to aluminum in
acid soils a review. Plant Cell, Tissue and Organ Culture 75: 189-207

Sarkunan, V., C.C. Biddapa, and S.K. Nayak. 1984. Physiology of aluminum toxicity in
rice. Curr. Sci. 53: 822-824.










Sharpley, A.N., and P. Kleinman. 2003. Effect of rainfall simulator and plot scale on
overland flow and phosphorus transport. J. Environ. Qual. 32: 2172-2179.

Sharpley, A.N., and R. Menzel. 1987. The impact of soil and fertilizer phosphorus on
the environment. Advances in Agron. 41: 297-324.

Shelton, C.H., R.D. von Bernuth, and S.P. Rajbhandari. 1985. A continuous-application
rainfall simulator. Trans. ASAE 28: 1115-1119.

Soil Science Society of America (SSSA). 2006. Soil science terms glossary [Online].
Available at http://www.soils.org/sssagloss/index.php. (verified 5/16/06). SSSA,
Madison, WI.

Soil Survey Staff. 2003. Soil Survey of Okeechobee County, FL. U.S. Gov. Print.
Office. Washington D.C.

South Florida Water Management District (SFWMD). 1997. Surface water
improvements and management plan Update for Lake Okeechobee. Planning
document. West Palm Beach, FL.

Swanson, N.P. 1965. Rotating boom rainfall simulator. Trans. the Amer. Soc. of Agric.
Eng. 8: 71-72.

Taylor, G.J., A.H. Adriano, and A.H. Johnson (ed.). 1989. Acidic Precipitation:
Biological and Ecological Effects. Springer Verlag, New York, NY. pp. 327-361.

United States Department of Agriculture (USDA). 1990. State soil geographic database
for Florida. U.S. Dept. of Agric., Soil Conservation Service, Gainesville, FL.

U.S. Environmental Protection Agency (US EPA). 1988. Test methods for evaluating
solid waste. Volume IC: Laboratory Manual Physical, Chemical Methods. 3rd ed.,
PB 8-239223, SW-846. Off. Solid Waste Emerg. Resp., Washington, D.C.

U.S. Environmental Protection Agency (US EPA). 1993a. Methods for determination of
inorganic substances in environmental samples. 356.1. Revision 2.0.

U.S. Environmental Protection Agency (US EPA). 1993b. Test methods for evaluating
solid waste, physical/chemical methods. Method 3050A. 3rd ed.

U.S. Environmental Protection Agency (US EPA). 1993c. Test methods for evaluating
solid waste, physical/chemical methods. Method 350.1. 3rd ed.

Van Wambeke, A. 1976. Formation, distribution, and consequences of acid soils in
agricultural development. In: Plant adaptation to mineral stress in problem soils
(Ed. M.J. Wright and S.A. Ferrari). Cornell Univ. Agric. Exp. Sta. pp. 15-24.







60


Withers, P.J.A., S.D. Clay, and V.G. Breeze. 2001. Phosphorus transfer in runoff
following application of fertilizer manure and sewage sludge. J. Environ. Qual. 30:
180-188.

Yu, B., C.A.A. Ciesiolka, and P. Langford. 2003. Calibration of oscillating nozzle-type
rainfall simulator. Earth Surf. Process. Landforms 28: 1483-1490.

Zhang, G.C., and G.J. Taylor. 1988. Effect of aluminum on growth and distribution of
aluminum in tolerant and sensitive cultivars of Triticum-Aestivum L. Comm. Soil
Sci. Plant Anal. 19: 1195-1205.
















BIOGRAPHICAL SKETCH

Thomas J. Rew ("TJ") was born on August 15, 1981. He is the first child of Tom

and Diana Rew. TJ grew up in a military family and attended eight schools before

earning his undergraduate degree from the University of Florida in Environmental

Management in Agriculture with a Land and Water Specialization (August 2003). While

a high school and undergraduate student, TJ spent his summers and holidays working on

the family cattle ranch, Hayman's 711 Ranch. It was here he developed a love of the

outdoors. This experience led TJ to develop an interested in the interdependency of the

environment, agriculture, and the world' s growing population. In August of 2003, TJ

began his master's work in soil and water science, focusing on phosphorus loss from

manure impacted soils, at the University of Florida.




Full Text

PAGE 1

ALUMINUM WATER TREATMENT RESID UALS FOR REDUCING PHOSPHORUS LOSS FROM MANURE-IMPACTED, HIGH-WATERTABLE SOILS By THOMAS J. REW A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Thomas J. Rew

PAGE 3

To my Granddad, my partner in crime, for a ll the summers he encouraged my passion for agriculture, the environment, and life

PAGE 4

iv ACKNOWLEDGMENTS I thank Dr. D.A. Graetz for his guidance a nd assistance over the past five years through both my undergraduate and graduate studie s. I also would like to thank the other members of my committee: Drs. V.D. Nair G.A. OÂ’Connor, and L.E. Sollenberger for their support and guidance. Special thanks go to the crew who helped collect my soil sample and prepare the rainfall simulation boxes: Jacob Butler, Eric Smith, Leighton Walker, and Steve Robinson. I would also like to thank Dawn Lucas fo r her help with laboratory experiments, calculations, and controlling emotions when using the AA; Julie, Lacey, Justin, and Ryan for their help in the lab (dishes); the CA LS Ambassadors and Agronomy and Soils Club for allowing me to become more than just a student of UF. In addition, I need to thank the faculty and staff of the University of Florida Soil and Water Science Department giving me the opportunity to expand my knowledge. Most importantly, I want to thank my family and friends for the support and advice they have given me over the past three years. Mom and Dad, IÂ’m finally done; Courtney Maibach, it is your turn to read my thesis and I promise, IÂ’ll fini sh arguing yours soon. Finally, Kelly Aynes, thank you for allowing me to setup an office in your house. I could not have finished without your love, patien ce, and understanding. I can only hope I can return love and support over the next two years of vet school. Thank you.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES............................................................................................................vii LIST OF FIGURES...........................................................................................................ix ABSTRACT....................................................................................................................... ..x CHAPTER 1 INTRODUCTION........................................................................................................1 2 LITERATURE REVIEW.............................................................................................6 Forms of Phosphorus in Soil.........................................................................................6 Water Treatment Residuals as Soil Amendments......................................................10 Aluminum Toxicity....................................................................................................14 Rainfall Simulation.....................................................................................................17 3 MATERIALS AND METHODS...............................................................................20 Project Description.....................................................................................................20 Site Description and Soil Collection...................................................................20 Aluminum Water Treatment Residual Collection...............................................21 Construction of Runoff Boxes.............................................................................21 Rainfall Simulation..............................................................................................23 Experimental Design...........................................................................................26 Soil Sampling and Forage Harvest......................................................................27 Analytical Methods.....................................................................................................27 Soil.......................................................................................................................27 Aluminum Water Tr eatment Residuals...............................................................28 Forage..................................................................................................................29 Runoff, Subsurface Flow, Leachate....................................................................29 Quality Assurance and Quality Control..............................................................31 Statistical Analysis..............................................................................................31 4 RESULTS AND DISCUSSION.................................................................................32 Initial Soil P Concentrations and pH..........................................................................32

PAGE 6

vi Aluminum Water Treatment Residual Characteristics...............................................32 Runoff, Subsurface Flow, and Leachate Characteristics............................................33 Volume................................................................................................................33 pH........................................................................................................................34 Phosphorus..........................................................................................................34 Nitrateand Ammonium-Nitrogen......................................................................42 Total Calcium, Magnesium, Iron, and Aluminum Concentrations.....................43 Preand Post-simul ation Comparisons.......................................................................44 Soil Phosphorus...................................................................................................44 Soil pH.................................................................................................................45 Forage......................................................................................................................... 46 5 CONCLUSIONS........................................................................................................51 LIST OF REFERENCES...................................................................................................53 BIOGRAPHICAL SKETCH.............................................................................................61

PAGE 7

vii LIST OF TABLES Table page 4-1 Initial soil mean values of P concentr ations and pH of the Immokalee fine sand used in the study.......................................................................................................32 4-2 Selected characteristics of the Manatee County Al-WTR.......................................33 4-3 Average runoff, subsurface flow, and leachate volumes as influenced by treatment from six simulations of runoff, five simulations of subsurface flow, and two simulations of leachate...............................................................................34 4-4 Average soluble P concentrations by trea tment in runoff (six events), subsurface flow (five events), and leachate (two events)...........................................................36 4-5 Average soluble P concentrations in r unoff, subsurface flow, and leachate after 4 wk of flooded conditions (Simulation 7)..................................................................39 4-6 Runoff, subsurface flow, and leachat e collected during simulation 7(A) Volume. (B) pH......................................................................................................40 4-7 Soluble P mass loss (SP concentratio n volume) per box averaged over six rainfall events and estimated SP lo ss per hectare per rainfall event........................41 4-8 Total, soluble, and particulate P conc entrations in runoff for each treatment averaged over six rainfall events..............................................................................42 4-9 Average N concentrations in runoff, subsurface flow, and leachate from six, five, and two simulations, respectively. (A) NH4-N. (B) NO3-N............................43 4-10 Average total Ca, Mg, Fe, and Al concen trations in runoff, subsurface flow, and leachate from six, five, and two simula tions, respectively. (A) Runoff. (B) Subsurface flow. (C) Leachate................................................................................44 4-11 Preand post-simulation water extracta ble P concentrations in the 0-10 and 1020 cm soil layers.......................................................................................................45 4-12 Preand post-simulation Mehlich-P concentrations in the 0-10 and 10-20 cm soil layers..................................................................................................................45

PAGE 8

viii 4-13 Preand post-simulation soil pH values in the 0-10 and 10-20 cm soil layers........46 4-14 Stargrass shoot biomass harv ested prior to each simulation....................................47

PAGE 9

ix LIST OF FIGURES Figure page 3-1 Runoff box design....................................................................................................22 3-2 Schematic diagram of rainfall simulator..................................................................24 3-3 Box placement under simulator................................................................................25 3-4 Treatment descriptions.............................................................................................26 4-1 Runoff, subsurface flow, and leachat e pH values from six, five, and two simulations, respectively..........................................................................................35 4-2 Average treatment SP concentrations a nd standard deviations from the three collection points.......................................................................................................38 4-3 Total shoot and root growth per tr eatment averaged over six simulations..............47 4-4 Average stargrass TP concentratio ns per treatment averaged over all simulations...............................................................................................................48 4-5 Average stargrass TN concentratio ns per treatment averaged over all simulations...............................................................................................................49 4-6 Average stargrass Al concentratio ns per treatment averaged over all simulations...............................................................................................................50

PAGE 10

x 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 ALUMINUM WATER TREATMENT RESID UALS FOR REDUCING PHOSPHORUS LOSS FROM MANURE-IMPACTED, HIGH-WATERTABLE SOILS By Thomas J. Rew December 2006 Chair: Donald Graetz Major Department: Soil and Water Science Dairy and beef operations in the Lake Okeechobee watershed in Florida and across the nation are receiving atten tion as a result of their c ontribution of phosphorus (P) to surficial water bodies. Numerous efforts ar e being made to suppor t the agricultural industry by reducing P losses from the soil. One such effort involves the addition of water treatment residuals (WTRs) to the soil. Prior research has s hown that Al-WTRs are capable of binding P and therefore reduci ng P loss through runoff and leaching. The objective of this research was to evaluate the effect of Al-WTR on P loss from a manureimpacted soil obtained from a dairy sprayfield using a rainfall simula tion protocol. Soil was removed from the field site as 0–10 and 10–20 cm depths. Both depths contained high concentrations of water-soluble P and Mehlich-1 P; approximately 18 and 950 mg P kg-1, respectively. After air drying and sievi ng, the soil was placed in rainfall simulation boxes (100 cm x 30 cm x 20 cm) designed to collect runoff, subsurface flow, and leachate. An Al-WTR was either surface app lied or incorporated to 10 or 20 cm depths

PAGE 11

xi at a rate of 2.5% of soil dry weight. Th e soil was then sprigged with stargrass ( Cynodon nlemfuensis ). Rainfall simulations were run six times at 3 wk intervals. Runoff was collected for 30 min after initial runoff be gan. Subsurface flow and leachate were collected (depths of 10 and 20 cm, re spectively) after runoff ceased. When Al-WTR was surface-applied, the SP concentration in runoff was reduced by approximately 75% compared to untreated soil ; however, SP concentrations in subsurface flow and leachate did not decrease. When Al-WTR was incorporated into the soil at depths of 0–10 or 0–20 cm, runoff SP concen trations were reduced by approximately 45%. Incorporation of Al-WTR to a depth of 10 cm decreased SP concentrations in subsurface flow and leachate by 37 and 11%, re spectively. However, with incorporation of Al-WTR to a depth of 20 cm, both subsur face flow and leachate SP concentrations were reduced by approximately 90%. The incorporated Al-WTR reduced soil waterextractable P (WEP) by approximately 70%. Ho wever, Mehlich-1 P concentrations were not affected by the incorporati on of Al-WTR in the soil. Care must be taken to ensure complete incorporation of Al-WTR throughout the P-impacted layer, as Al-WTR is only effective in reducing SP concentrations when it is in contact with the impacted soil. Shoot and root growth of stargrass were not adversely affected by the Al-WTR applied at a rate of 2.5% of soil weight.

PAGE 12

1 CHAPTER 1 INTRODUCTION Over the past 5,000 yr, the Lake Okeechobee watershed has developed an ecosystem to thrive under oligotrophic c onditions. The watershed is approximately 12,000 km and flows roughly from Orlando to the Everglades (SFWMD, 1997). On average, the watershed receive s 120 cm of rain annually (M cCaffery et al., 1976). The elevation in the watershed is predominatel y flat, with a range from 3.1 to 22.9 m above mean sea level (Soil Survey, 2003). As a result the watertable is na turally close to the surface for much of the year. Prior to deve lopment and utilization of the land, much of the watershed was completely saturated fo r one to two months a year (Gatewood and Bedient, 1975). The watershed is divided into six regions. The two most import ant regions are the Lower Kissimmee River (LKR) and Taylor Creek/Nubbin Slough (TCNS) because they contribute approximately 57% of the tota l phosphorus (P) load to Lake Okeechobee (Fluck et al., 1992). Phosphorus discharge to surficial water bodies has recently become a major concern in the watershed. One of th e current major contributors to the situation is waste from livestock which contains high amounts of P. As Florida has continued to develop, land has become more valuable and farmers and ranchers have intensified their operations by increasing the concentration of animals per acre. The resulting effect has been increased discharge of P into the watershed. Because of the warm climate and abundance of water, the agricultural community has flourished, and land use practices in th e LKR and the TCNS have intensified to

PAGE 13

2 maximize production. In particular, the beef and dairy industries have, respectively, amassed herds of 133,000 and 32,000 animals in Okeechobee County (Florida Agriculture Statistical Direct ory, 2004). To support these he rds, 31% of the land in the LKR and 62% of the land area in the TCNS has been drained and converted to improved pastures. Improved pasture area increased by 500% between 1950 and 1970. Currently, improved pastures comprise approximately 87,000 ha in the LKR and the TCNS (Flaig and Havens, 1995). Inorganic fertilizer used on improved pastures account for 34% of all P imports to the watershed (Boggess et al., 1995). Additional P imports come from mineral supplements and winter feed. The mo st intensive land use in the watershed is dairy farming. From 1960 to 1990, dairies increa sed their imports of P in cattle feed from 360 t P yr-1 to 1200 t P yr-1 (Flaig and Havens, 1995; Boggess et al., 1995). Dairy feeds constituent 35% of P imports into the wate rshed, but P impacts have been decreasing since 1990 as a result of the dairy buyout progr am which closed 19 of 49 dairies in the basin (Flaig and Reddy, 1995). Phosphorus is a vital nutrient in both natu ral and agricultural e nvironments and is often imported to foster agricultural operati ons. Traditionally, P has been considered relatively immobile in soils, primarily only subject to loss by erosion. However, P can also move vertically or horizontally within some soil profiles and can be transported into the surface water of canals, streams, and lake s. Graetz and Nair (1995) found residual P from manure has the potential to leach laterally because of the surface horizonÂ’s low Pretention capacity in a Spodosol. The resul ting soluble P in surface water often causes detrimental impacts on native ecosystems.

PAGE 14

3 Phosphorus occurs in soluble and in soluble compounds in both organic and inorganic forms in the soil environment (Fau lkner and Richardson, 1989). Organic forms exist as both living plant tissue and in deposits of detrital material. Inorganic P is found as insoluble phosphate and minerals adsorbed to clay particles, or ganic matter, and Fe and Al hydroxides (Mitsch and Gosselink, 1993). Nevertheless, the majority of P is unavailable to biota. Unlik e nitrogen, P does not have a ny significant losses to the atmosphere. Therefore, P remains in the so il or is transported by another mechanism besides volatilization. Soils in the watershed consist predominatel y of Spodosols, Entisol s, and Histosols. The Histosols occur primarily as small deposits in wetlands, where as Spodosols dominate the northern portion of the wate rshed. The surface horizons are typically between 0.08 and 0.20 m thick and are underl ain by a spodic horizon at 0.5 to greater than 2 m depth (USDA, 1990). The closer the spodic horizon is to the surface, the more poorly drained the soil. Spodosols are charact erized by high infiltration rates because of the greater than 90% sand content. However, they have poor internal drainage because of the low permeability of the spodic horizon. The Spodosols, Myakka (32%), Immokalee (30%), and Pomello (2%) cove r 64% of Okeechobee County (So il Survey Staff, 2003). The concern regarding the high watertable a nd high infiltration ra tes is that surface horizons of Entisols and Spodosol s have a limited capacity to retain P and, thus, are susceptible to P loss. The Bh (spodic horizon) in South Florida Spodosols contains Al and/or Fe and organic material accumulated from the stripping of the E horizon. Spodosols in general are naturally deficient in P (Hodges et al., 1967). Neverthe less, as a result of the

PAGE 15

4 accumulation of material, the spodic has the ab ility to retain downward moving P onto Fe and Al oxides (Nair et al., 1995). These eluvia l materials enable the possible retention of excess P from agricultural operations. De pth to Bh has an impact on the P leaching potential of the soil. As the depth of th e Bh increases the potential for P loss via subsurface drainage increases. Moreover, as the depth of the spodic increases its P adsorption capacity decreas es (Nair et al., 1999). Despite the sorption of P in the soil, it can still be highly mobile. The combination of the fluctuating and perched watertable a llows lateral movement of P through the soil and into the surface water (Graet z and Nair, 1995). In fact P can move so readily, that between 1973 and 1988, the concentration of P in Lake Okeechobee increased approximately 250% (Neganban, 1993). In an effort to reduce eutrophication a nd restore the Everglades to a system resembling the past, larger rel eases of water to the Everglad es and reductions in P loading are mandatory. Therefore, to meet the re quirements of the Comprehensive Everglades Restoration Plan, the South Fl orida Water Management Dist rict, U.S. Army Corps of Engineers, and other state agencies created massive water treatment areas, instituted use of Best Management Practices (BMPs), and increased water flow to the Everglades (SFWM, 1997). Representative BMPs include fencing off canals, adding water troughs to pastures, and reducing P c oncentrations in feed. These measures alone are not adequate to meet the acceptable P levels within the watershed due to residual P levels remaining in soils highly impacted by manure. Recent studies have suggested that th e addition of soil amendments has the potential to reduce P loss from soil. Investigations have in cluded numerous amendments from differing

PAGE 16

5 sources. One promising area of investigation is the land applicati on of water treatment residuals (WTRs). WTRs ar e the waste product from drinking water purification, and typically contain Al, Ca, or Fe compounds us ed to remove nutrients and particulate matter from source water. Land application of WTRs has the potential to help reduce water soluble P (WSP) concentrations in P-imp acted soils. Previous research has shown that WTRs containing Al are the most effici ent in reducing WSP in soils. Therefore, a study was developed to evaluate the effect of an Al-WTR on soil P dynamics in a sandy soil with a fluctuating watertable. The study had two main objectives: 1. To determine the effect of surface applic ation, partial mixing, and complete mixing of Al-WTR with soil on P loss in runoff, subsurface flow, and leachate. 2. To determine the Al-WTRÂ’s effect on forage growth.

PAGE 17

6 CHAPTER 2 LITERATURE REVIEW Forms of Phosphorus in Soil Phosphorus (P) is commonly found in soils in small total qua ntities (Lindsay, 1989). Forms of P in soils can be either i norganic or organic. Phosphate minerals, organometallic complexes, and P to bound meta ls and hydroxides are all inorganic forms of P that may be present in the soils. Microbes, humus, and undecomposed litter make up the organic P fraction (Gale et al., 1994). Reddy et al. (199 6) identified five major pools of P in soils: a) labile inorganic P, b) P bound to Fe and Al minerals, c) P bound to Ca and Mg minerals, d) P bound in labile orga nic forms, and e) residual organic P. The labile pool of P is of particular environmental concern because labile P can potentially move laterally or ve rtically in the soil profile as a result of precipitation. The Glossary of Soil Science Terms (SSSA, 2006) defines the labile pool of P as “that portion which is readily solubilized or exchanged when the soil is equilibrated with a salt solution,” and the available pool as “the amount of soil P in chemical forms accessible to plant roots or compounds likely to be conve rtible to such forms during the growing season.” Oxalate-extractable Al and Fe dominate P retention in many soils and provide strong sorption of P (Ballard and Fiskell, 1974 ). The retention of P by Fe/Al or Ca/Mg depends on the soil’s pH. In acid and neutral soils, P retention is dominated by Fe and Al compounds because P binds with protonated su rfaces of hydroxides of Fe and Al (Olila and Reddy, 1995). In alkaline soils, Ca and Mg compounds are more dominant in

PAGE 18

7 retaining P because P sorption decreases due to competition between PO4 3and OHanions. Anion exchange binds phosphates wi th Fe and Al hydroxides (Olila and Reddy, 1995). Because amorphous Al and Fe hydr oxides have a larger number of singly coordinated surface hydroxyl ions, they have a higher P sorption capacity than crystalline oxides (Reddy and Smith, 1987). In reduced conditions, Fe stability is affected. As a result of the reduced stability, P associated with Fe3+ maybe solubilized and released into the soil solution. However, Al compounds asso ciated with P are not affected by changing oxidation and reduction c onditions in the soil (Miner, 2001). Labile organics represent the short term stor age of P as living tissue in plants. This P is quickly returned to the soil eith er through manure or decomposition, whereas, residual organics represent long term storage of P as detrital tissue that is resistant to decomposition (Reddy et al., 1996). Because of the P storage ability of plant tissue, phytoremediation of warm-season perennial gr asses has been examined. Warm-season perennials have the ability to extract more P than warm-s eason annuals and cool-season grasses because the po tential dry matter yield of perenni als is greater. Howard (2006) estimates that adding nitrogen (N) to increa se forage yields in hay production could decrease soil P concentration 4 to 13 mg kg-1 annually. The soilÂ’s ability to retain P determines the risk of environmental consequences related to P leaching. Therefor e, the continued assessment of a soilÂ’s capacity to retain P remains a vital tool in environmental sustai nability. Soluble forms of P react and form less soluble compounds with soil components. The amount of P available to be removed from solution depends on the quantity of P -reactive colloidal surfaces within the soil (Holford et al., 1997). The sorption and de sorption capacity of soils between horizons

PAGE 19

8 has the potential to influence the movement of P within the prof ile (He et al., 1999). Variations of P sorption are attributed to surface area variability based on particle size (Atalay, 2001). Soil sorption and desorption mechanisms control the solution phase P (Frossard et al., 1995). Determining the movement and availabil ity of P in soils and groundwater is paramount in the preservation of natural res ources. By using the Langmuir adsorption isotherm equation, an estimate of P sorption capacity and strength can be determined in the soil. Soil extractable P can be related to the P-sorbing capacity of a soil by using the degree of P saturation test (DPS); DPS = (ext ractable soil P / P so rption maximum) 100 (Breeuwsma and Silva, 1992). Degree of P saturation may also be expressed as a percentage of double-acid extractable P (Mehli ch-1) to the P sorption capacity of a soil (Nair and Graetz, 2002). When sorption equals desorption a system is at equilibrium. Therefore, if the concentration of P in soil water decreases, desorption will occur until equilibrium is reached. Likewise, if the P in soil water in creases, sorption will occur until equilibrium is attained (Froelich, 1988). The occurrence of no net sorption or de sorption is known as the equilibrium P concentration (EPC). Sharpley and Menzel (1987) found sorption properties of the soil control the conversion fr om dissolved P to particulate P and back to dissolved P. Therefore, the concentration of P controls the direction of exchange. A prediction of loss or gain of P from solution can be determined using the EPC; soils with low EPC values tend to sorb soluble P. Graetz and Nair (1995) found that the A and E horizons of a manure-impacted Spodosol had higher EPC values than the Bh horizon. Therefore, A and E horizons are

PAGE 20

9 inferred to have a lower retention capacity than the Bh horizon. In manure-impacted soils, high EPC values reflect high P loadi ng. Graetz and Nair (1995) also found that oxalate-extractable Al and to tal organic carbon accounted for approximately 69% of the variability in the P retention maximum in the A, E, and Bh horizons forage, pasture, and intensive areas. Maximum P retention capacity increases with depth. In fact, the Bh was found to have a P retention capacity three to four times greater than the A and E horizons (Reddy et al., 1996). The high cost of inorganic N fertilization ha s prompted agriculture to better utilize waste products through the application of biosolids and animal manures. These amendments are typically app lied to meet a crop’s N require ment. The result of this common practice is an excessive accumulation of P in the soil. Soils with low P sorption capacities, such as sandy soils, are more susceptible to P loss through runoff and subsurface water movement. Kleinman and Sh arpley (2003) evaluate d P runoff from two soils amended with three types of manure (applied at 6 rates ranging from 0–150 kg TP ha-1) following the National P Protocol for rainfa ll simulation studies. Dissolved reactive P (DRP) in runoff increased with increased ma nure application rate. Water-extractable P (WEP) concentrations in the ma nures were directly related to DRP runoff concentrations. Repeated rainfall events diminished DRP con centrations in runoff with all manure types and application rates. This trend was attribut ed to both the transloc ation of manure P into the profile and the loss of prev ious applied P by runoff. A close association between soluble P (SP) concentration in land-applied manures and P concentrations in runoff has been shown in recent studies. Moore et al. (1994) and Withers et al. (2001) indicated that DRP lo ss from amended soils was proportional to the

PAGE 21

10 soluble P content of the source. In a ddition, Kleinman et al. (2002) found the WSP concentration of surface applied manure (dair y, swine, and poultry) was highly correlated with DRP in runoff from three soils. Water Treatment Residuals as Soil Amendments Efforts to reduce P loss in agricultural sandy soils include several strategies: reducing P loading, increasing the ambient P storage, and maximizing P retention. Many soils with high P concentrations currently discharge P into su rficial water bodies. Phosphorus retention in highly impacted sandy soils could be improved with the addition of amendments such as water treatment residuals (WTRs) de rived from the treatment of drinking water to remove color, taste, turb idity, and odor. The chemical composition of WTRs depends on the metal salts used to clarif y the water. The typi cal salts used are Al (alum), Fe (ferric chloride), or Ca (CaCO3). The by-product of this water purification process is solid material commonly referred to as WTR. The WTR is often considered a waste product because it is currently being dis posed in landfills, sanitary sewers, or in lagoons (Ippolito et al., 2002). Water treatm ent residuals have hi storically had little success as a soil substitute or soil amendmen t compared to their counterpart, biosolids (Cornell and Westerhoff, 1981). However, recently WTRs have been considered as soil amendments because of the P-sorbing characteristics. Recently WTRs have been evaluated for use as both a soil substitute and a soil amendment because their properties are similar to fine textured soils (Elliott et al., 1990). These fine textured substances are comprised of sand, silt, an d clay particles as well as activated carbon, polymers, Al, Fe oxides, a nd calcium carbonate derived from the raw water (Elliott and Dempsey, 1991). The WTRs have improved soil conditions in numerous studies by increasing organic matter content, water holding capacity, and pH.

PAGE 22

11 However, because of the high relative surface area of the amorphous Al and Fe compounds, Aland Fe-WTRs can cause P defici encies in soils and reduce crop yields (Bugsbee and Frink, 1985). Dayton and Basta (2001) tested the benefi cial properties of WTRs by evaluating 17 WTRs for use as a soil substitute by comparing their nutritive, physical, and chemical properties with soil. The WTRs contained the full spectrum of available nutrients, except P. A bioassay was performed with tomato seedlings ( Lycopersicon esculentum ) to validate the results of soil tests used to meas ure P adequacy in the WTRs. All plants had low tissue P (561–1840 mg kg-1; median 923 mg kg-1); 1000 mg kg-1 is considered deficient. Vegetative yield was limited primar ily because of P deficiencies and in some cases, phytotoxic levels of NO2-N ( 10 mg kg-1). In an effort to utilize the positive as pects of organic wastes and minimize detrimental effects, Gallimore et al. (1999) examined four surface application treatments of Al-WTR (0, 11.2, 44.9 Mg ha-1 and 44.8 Mg ha-1 in a buffer strip at the end of the plot) on bermudagrass plots receiving 6.72 Mg ha-1 of poultry litter. The plots received simulated rainfall for 75 mins at 6.3 cm hr-1 within 24 hr of the a pplication of the litter and Al-WTR. Surficial runoff was collected and analyzed for TN, NH4-N, TP, SP, dissolved Al, and dissolved solids. No re ductions in SP concentrations were observed with the 11.2 Mg ha-1 rate. Soluble P concentrati on was reduced from 15.0 mg L-1 in the control to 8.6 mg L-1 in the 44.9 Mg ha-1 application rate. Similar results were obtained when the WTR was applied just in the buffer strip. These reductions were attributed to the amorphous Al in the Al-WTR. In additi on, dissolved solids and Al content did not increase in surface runoff with the application of the Al-WTR.

PAGE 23

12 The potential P-sorption and precipitation capability of Al-, Ca-, and Fe-WTRs has generated interest in their use as an amen dment in highly P-impacted soils. OÂ’Connor et al. (2002) found that WTRs reduced P solubil ity and leaching in manure-amended soils. Brown and Sartain (2000) re ported that Fe-WTR was able to significantly reduce P leaching on a simulated golf course using 2.5% by weight of the Fe-WTR. Al-WTRs reduced P and NH4 + by 75% in surface runoff from land treated with poultry manure (Basta and Sloan, 1999). All soluble P was ad sorbed in an 8:1 mixture of Al-WTR and biosolids (Fort Collins, CO). Increasing this ratio has the potential to adsorb all available P in the biosolids and the soil P (Ippolito et al., 2002). A rainfall simulation runoff study by Hauste in et al. (2000) evaluated both Al-WTR and HiClay Alumina (HCA-a byproduct of co mmercial alum production) to test P adsorption capabilities in a soil highly imp acted by P. The HCA P-adsorption capacity was 20 times less than the Al-WTR Padsorption capacity (86 vs. 1750 mmol kg-1). In addition, the Al-WTR increased the total rec overable Al in the so il, while HCA did not affect the recoverable Al concentration. Thes e differences were attributed to the greater total Al content of the Al-WTR co mpared to the HCA (159 vs. 46.7 g kg-1) The effect of WTR composition on P sol ubility and leaching was tested by Elliott et al. (2002) in a sandy soil (I mmokalee-Spodosol) after additions of biosolids and triple superphosphate (TSP) to increase P concentra tions in the soil. The study evaluated the ability of Al-WTR, Fe-WTR, Ca-WTR, and pur e hematite to reduce P loss from the soil. Soluble P concentrations in the leachate de creased in the order: Al-WTR > Ca-WTR Fe-WTR>>hematite.

PAGE 24

13 The retention mechanisms of an Al-WTR were studied by Ippolito et al. (2003). The research involved shaking an Al-WTR for 1–211 d and analyzing the solution for pH, Ca, Al, and P. The pH increased from 7.2 to 8.2 after shaking. Change in pH beyond 84 d of shaking was not significant. In addition, the shaking increased Ca and Al concentration and decreased P concentrati on. The average maximum pH of 8.15 and Ca desorption suggest that the Al-WTR’s Ca sour ce was the raw water used at the treatment plant (South Platte Rive r pH~7.5–8 and Ca = 41.9 g kg-1). Removal of P was attributed primarily to chemisorption on the amorphous Al mineral phase. Novak and Watts (2004) evaluated the im pact of incorporating two Al-WTRs on two Ultisols in a laboratory setting. The Al -WTRs were added at the following rates: 2.5, 5.0, 7.5, and 10.5% by weight. The Pmax values for both amended soils (175 and 85 mg P g-1) were significantly higher than Pmax values without the addition of Al-WTRs ( 1.0 mg P g-1). The results demonstrate the usability of Al-WTRs to increase P adsorption in sandy soils. In addition, the fi ndings suggest off-site P-transport could be reduced by the implementation of this new chemical based BMP. Application rate of an Al-WTR (0, 2.5, 5.0, and 10.0% by weight), the degree of mixing of the Al-WTR with th e soil, and the mixing of an un-impacted E-horizon with an impacted A-horizon on P leaching was examined by Miyittah-Kporgbe (2004). Runoff and leaching rates were determined via rain fall simulations and columns, respectively (leachate was also collected from rainfall si mulation.). Phosphorus leaching was reduced by 87 to 99.7% when the Al-WTR was mixed co mpletely with 15 cm of soil, compared to partial mixing (7.5 cm of soil), which only reduced P leaching from 40 to 58%. Improved P retention was attributed to increase d contact of the Al-WTR with soil P. The

PAGE 25

14 runoff results found that soluble P concentr ations met the cri tical threshold of 0.03 mg L-1 when Al-WTR was surface applied. Howe ver, due to limited contact between the amendment and soluble P, the leachate solubl e P concentrations exceeded the critical value. Combining the impacted A-horizon with the un-impacted E-horizon increased the Al-WTR efficiency. This increase in efficien cy was attributed to the dilution of soluble organic material, which can block P adsorp tion sites on the Al-WTR. Conclusions from the data suggest Al-WTR must be in direct contact with soluble P to prevent loss by leaching. In addition, an increase in surface application (10%) can be as effective as mixing lower quantities (2.5%) of Al-WTR. Application and distribution of Al-WTR in the soil must be made in accordance with an ticipated P losses; surface application with runoff or mixing with leaching. Aluminum Toxicity Aluminum toxicity to plants is a primary concerns when Al-WTR is applied to soil. In acidic soils, Al toxicity has been show n to be one of the primary growth-limiting factors (Foy et al., 1978). Mob ility and phytotoxicity of Al in the soil solution as a result of soil acidification (Taylor et al., 1989) is a potential problem. Soil acidity impacts approximately 40% of the arable soil in the world (Haug, 1984). As a result, Al phytotoxicity presents a serious agricultural concern (Van Wambeke, 1976). Symptoms of Al toxicity, such as reduced root a nd shoot growth, mimic phosphate, Ca, and Fe deficiencies (Foy, 1984). Extens ive Al research has been conducted which indicates Al toxicity primarily affects the roots: a) the observable symp toms of Al toxicity are the reductions in root elongation (Osborne et al., 1981; Jarvis and Hatch, 1986); b) the production of root biomass compared to shoot biomass is typically more sensitive to Al

PAGE 26

15 (Buss et al., 1975; Zhang and Tayl or 1988); c) in Al-stressed pl ants there is no correlation between an accumulation of Al in the shoots and the supply of Al (Foy et al., 1972); and d) the Al tolerance of a cultivar has been shown to be determined by the rootstock in grafting experiments (Klimashe, 1970). The visual appearance of mature roots is not the exclusive de termination of Al toxicity or tolerance. At the cellular leve l, Al influences leaf tissue metabolism (Ohki, 1986; Hoddinott and Richler, 1987; Sarkunan et al., 1984). Tayl or et al. (1989) suggested potential mechanisms of Al disruptio n on cell functions to be: a) disruption of membrane structure and functions; b) inhi bition of DNA synthesis and mitosis; c) inhibition of cell elongation; d) disruption of mineral nutrition; and e) disruption of phosphate and Ca metabolism. Taylor also suggested several possible immediate toxic effects of Al. The membrane structure and/ or function at the soil root interface maybe altered. Membrane-bound enzymes will be affected. Aluminum may also reduce cell elongation by affecting cell wall component s or assembly. Phosphate in the DNA appears to bind with Al in the cytosol. As a result, DNA synthesi s will be inhibited because of repressed template activity. Finally, toxic effects of enzyme-mediated reactions in phosphate metabolism seem probable. Aluminum occupies approximately 7% of the of earthÂ’s crust. Based on the toxicity issues described prev iously, it is fortunate the ava ilability of Al is reduced by ligands or the Al occurs in other nonphytotoxic forms (i.e., aluminosilicates and precipitates) (Delhaize and Ry an, 1995). However, the solub ilization of Al is enhanced by low pH. Trivalent Al (Al3+) dominates in acidic conditions (pH<5), compared to Al(OH)2+ and Al(OH)2 + that form as pH increases. Near neutral pH, the solid phase

PAGE 27

16 gibbsite occurs (Al(OH)3) while in alkaline c onditions, aluminate (Al(OH)4 -) forms. Many of the monomeric Al cations have the ability to bind with organic and inorganic ligands, including: PO4 3-, SO4 2-, F-, organic acids, proteins, and lipids (Delhaize and Ryan, 1995). Since Al3+ is predominately restricted to acidic conditions and many trivalent cations are toxic to plants, it is typically a ssumed to be the major phytotoxic species. However, based on the complex nature of Al, th is has been difficult to prove conclusively (Delhaize, 1995). In addition, most of the monomeric Al species previously listed have been considered toxic in one or more studi es (Kinraide, 1991). To combat toxicity, numerous strategies have been utilized. The method preferred in North America and Europe has been the application of calcium car bonate (lime) to rais e soil pH. As the pH rises, the Al is converted to less toxic forms (Samac, 2003). Aluminum toxicity and heavy metal cont amination concerns have been raised regarding the land app lication of Al-WTRs. Dayton a nd Basta (2001) tested 17 WTRs (14 were alum based) for heavy metals and nutrients according to the U.S. Environmental Protection AgencyÂ’s (EPA) toxicity characte ristic leaching procedure (TCLP), a test designed to determine the mobility of organic and inorganic analytes in liquids, solids, and multiphasic wastes (U.S. EPA, 1988). The results of the test found that all of the WTRs contained significantly lower than re gulatory levels of the EPAÂ’s TCLP and the residuals were therefore, classified as nonh azardous wastes. Total N ranged 1.3 to 18.4 g kg-1 with an average value of 7 g kg-1. Typical soil levels of soil total N range from 0.2 to 5.0 g kg-1 (Dayton and Basta, 2001). The sources of higher N levels in the WTRs are the algae, detritus, etc. removed from raw water. Mineralization of organic N in the WTRs

PAGE 28

17 may pose potential problems related to NO2 concentration in WT Rs containing higher than 10 g kg-1 total N. Most significantly, Dayt on and Basta (2001) found soluble Al levels at a median concentration of 0.054 mg L-1, with a range from 0.02 to 0.92 mg L-1. Therefore, problems stemming from both Al t oxicity and heavy metals are not expected with WTR use. Elliot and Depsey (1991) found that unless the raw water source is contaminated, the nutrient conten t of WTRs is generally low. As a result, the WTRs pose only a minor threat to the environment. Rainfall Simulation Rainfall simulation studies over the past 40 yr have become popular because irregular distribution of rainfall hampers the possibility of reasona ble time periods for study (Neff, 1979). Simulators were first us ed for erosion studies. The two primary issues were raindrop size di stribution and energy (Esteves 2000). Through a variety of research, two types of rainfall simulators ha ve emerged: (i) drip formers (Farmer, 1973; Romkens et al., 1975; Munn and Huntington, 1976) and (ii) nozzles (Meyer and McCune, 1958; Swanson, 1965; Miller, 1987; Riley and Hancock, 1997). Pressurized nozzle systems have become the preferred method for large area field studies (10 to 500 m2) (Esteves et al., 2000). Simplicity and speed are the basic requirements for the movement and assembly of a rainfall simulator from one research location to another. As a result, compromises have been made between technical constraint s and the reproduction of natural rainfall characteri stics (Esteves et al., 2000). A critical factor for experiments using ra infall simulators is the estimation of rainfall at the ground level (Yu et al., 2003). As rain gaugeÂ’s catchment area is elevated above the groundÂ’s surface, the pressure of si mulated rain from the nozzle decreases as greater distances are reached. The elevat ed catchment area has the possibility to

PAGE 29

18 overestimate rainfall volume. Overestimation without adjustment could range from six to 29 percent depending on catchment height (Y u et al., 2003). Additional variability can occur due to water pressure differences nozzle aging, and imperfect nozzle spray overlap. Wind can also cause non-uniform spatial variation. Measures to avoid variability include better pre ssure control, stainless stee l nozzles, and the use of a windbreak (Yu et al., 2003; Esteves et al., 2000). An additional challenge for research is th e comparability of experimental results from different researchers. The potential for human error is enhanced by differing sampling techniques, recording methods, measur ements, and setup. To avoid differences in results, the Nationa l Phosphate Research Project (NPRP, 2001) adapted a specific design for rainfall simulators. The establishment of a single design for a simulator expedites data collection, promotes comparab le results, and attempts to maintain field relevancy (Sharpley and Kleinman, 2003). The designs of Shelton et al. (1985) and Miller (1987) are the basis of the portable rainfall simulato r used by the NPRP (Humphry et al., 2002). Numerous studies across the nation have used the NPRP protocol for rainfall simulations both for field and box studies to evaluate P loss under differing soils and amendment treatments. Kleinman et al. ( 2003), Moore et al. (2000) and Withers et al. (2001) have used rainfall simulators to de monstrate P loss in runoff. As our knowledge base has increased with these studies so has th e need with the design of the rainfall boxes. Miyittah-Kporgbe (2004) has used NPRPÂ’s r unoff box design and added a false bottom to collect leachate and examine WTRÂ’s effect on both runoff and leachate. Further studies

PAGE 30

19 are needed to investigate the impact on sol uble P with the mixing of WTRs at differing depths and the impacts on plant growth and chemical composition.

PAGE 31

20 CHAPTER 3 MATERIALS AND METHODS Project Description Site Description and Soil Collection Soil for this study was obtained from a tile-d rained sprayfield at Larson Dairy Barn 5 near Okeechobee, Florida (N 2716.088’, W 8046.460’). The field had received applications of dairy manure for several year s. In addition, effluent from the farm’s lagoon was applied to the field periodically as needed to maintain lagoon capacity. The soil was an Immokalee fine sand (sandy, siliceous, hyperthermic Arenic Alaquods) brought into agricu ltural production as a pasture. The site was planted with stargrass ( Cynodon nlemfuensis ) for silage production in 1990. A series of surface soil samples was taken over the entire field to select a sa mple site with relatively high P levels. These samples were analyzed for water-soluble P using a HACH Orthophosphate Test Kit and areas of high P levels were identified. Based on these results, an area within the field was selected and a more intensive grid sampling was conducted to identify the final sampling site. These samples were then analyzed fo r water soluble P (WSP) using a 1:10 soil to solution ratio and Mehlich-1 ex tractant (Mehlich, 1953) to determine P concentration. Bulk soil samples were obtained from 0–10 cm and 10–20 cm depths at the selected site in a 6.5 x 1.5 m area after rem oval of above-ground vegetation. The soil was transported to Gainesville, FL, air dried, and sieved (0.64 cm mesh) to remove debris and to homogenize the soil. Five samples from th e bulk dried and sieved soil were randomly

PAGE 32

21 collected per depth (10 in to tal) and analyzed for WSP a nd Mehlich-1 P to evaluate uniformity of the soils prior to packing the rainfall simulator boxes. Aluminum Water Treatment Residual Collection The aluminum water treatment residual (A l-WTR) for the study was obtained from the Manatee County water treatment plant in Bradenton, FL. Slurry ponds were used to hold the Al-WTR until it was dry enough to be moved with a front-end loader. The solid material was then moved to spoil piles to cont inue to dry before it was disposed of at a landfill. The Al-WTR samples used for this study were collected from the spoil piles. Size distribution of the Al-WTR particles varied widely. Material ra nged from clay sized particles to 10 cm fragments. The Al-WTR was air dried for 2 wk under an open-sided greenhouse, because of its high moisture conten t, before sieving to pass a 0.64 cm mesh. Even after the material was dried and sieved, the moisture content was 29% because of the high organic content of the Al-WTR. Five samples were randomly collected from the bulk Al-WTR and analyzed for WSP, Mehlich-1 P, and TP. Construction of Runoff Boxes Runoff boxes (100 cm long by 30 cm wide and 20 cm deep) were constructed using pine lumber (Fig. 3-1). A 4 cm rail was a dded to raise the bottom of the runoff box. A 2 cm lip surrounded the box on three sides to prevent loss of water through runoff and/or splashing. One end of the box was 2 cm lower to allow runoff to be collected. The box corners and sampling ports were sealed using latex caulk, and the entire box was covered with a gel coat to make the box watertight. Ports used to collect subsurface flow and leachate were covered with a double layer of ga rden weed screen to prevent blockage of the ports with debris. Thir ty-nine kg of the sieved 10–20 cm depth and then 39 kg of 0– 10 cm depth were added to replicate field soil depths.

PAGE 33

22 A slit PVC pipe (5.6 cm ID by 40.6 cm length) with e nd caps was attached to one end of the box to collect runoff. Gravity a llowed the runoff water to be collected through a port in the center of the pipe. Six drai nage ports with attached Tygon tubing were installed on the bottom of the boxe s to monitor the watertable a nd to collect leachate (Fig. 3-1). The ports were located in pairs: 77.0, 38.5, and 5.6 cm from the runoff collection end and 5.6 cm from the side walls. In a ddition, two ports with attached Tygon tubing were placed at the end of the box between the two soil layers to collect subsurface flow. Figure 3-1. Runoff box design. 1 1. Runoff collection point. 2. Subsurface drainage points. 3. Leachate drainage points. 2 3

PAGE 34

23 Soil was placed in the boxes either as discrete 0–10 cm (39.5 kg) and 10–20 cm (39.5 kg) depths or as a mixed layer at approximately 1.3 g cm-3 to mimic field conditions as noted in the experimental design secti on below. Al-WTR from Manatee County was applied to the appropriate treatments at a ra te of 2.5% of dry soil weight (0.975 kg of AlWTR dry weight per 10 cm of soil treated). The Al-WTR and soil were mixed using a cement mixer to evenly distribut e the Al-WTR and/or soil in the appropriate boxes. After establishing the appropriate treatments, stargr ass cuttings obtained from an area adjacent to the sampling site in the spra yfield were sprigged at 30 pe r box to simulate field density of grass. Simulation boxes were kept in an open-sided greenhouse during the study to protect them from rainfall. Rainfall Simulation Rainfall simulation was conducted accord ing to a protocol developed by the National Phosphorus Research Project (NPRP) to quantify soil P-runoff relationships (NPRP, 2001). Rainfall intensity wa s uniformly applied at 7.1 cm h-1 (approximately 210 cm sec-1) from a height of 3 m above the soil surface. This is equivalent to a 10 yr, 24 hr rain. Gainesville municipal water was adju sted from a pH of approximately 8.5 to 5 using 3 M HCl to mimic rainfall pH in South Florida. Rainfall was dispersed uniformly using a TeeJetTM HH-SS50WSQ nozzle centered under a 3 x 3 x 3 meter frame protected from wind (Fig. 3-2). Thirty c ups were distributed in a grid to test rainfall distribution uniformity. There were no signi ficant differences in volume within the grid. Rainfall simulations were repeated every 3 wk, for a total of 6 rainfall simulation events. Analysis of the source water and rainfa ll (pH adjusted source water) was conducted by inductively coupled plasma atomic emi ssion spectroscopy; total Fe and Al were approximately 0.15 and 0.45 mg L-1 respectively. Source water and rainfall also

PAGE 35

24 contained Ca (14 mg L-1) and Mg (8 mg L-1). These Ca and Mg levels were expected based on the municipalityÂ’s we ll field pumping water from the Floridan aquifer which passes through both limestone and the Hawt horne formation. In addition, the municipality treats the water with lime. Soluble P, TP, NO3 -, and NH4 + were below detectable limits (0.03, 0.03, and 0.02, and 0.06 mg L-1, respectively). Figure 3-2. Schematic diagram of rainfall simulator (NPRP, 2001). Rainfall boxes were placed five at a time under the simulator at a 1 to 2% slope (See Figure 3-3). Thirty minut es of runoff (collection poin t out side of simulator) was collected from the boxes via Tygon tubing conne cted to the runoff port. Runoff did not start at the same time for each box; therefor e each box was covered after 30 min of runoff had been collected. The boxes were left in place under the simulator for approximately half an hour while subsurface flow and leachate were collected. This was done by

PAGE 36

25 draining the two side and two front leachate po rts, respectively. All collected samples were weighed and then sub-sampled. A 1 L sub-sample was taken from runoff, and 250 mL sub-samples were taken from subsurface flow and leachate. Two filtered (0.45 m) sub-samples (20 mL) were taken from each s ub-sample, using a vacuum pump to obtain approximately 40 mL aliquot for analysis. The sub-samples were refrigerated until P analysis was performed, usually the following day. One 20 mL filtered sample was acidified and refrigerated until NO3 and NH4 + could be measured. Figure 3-3. Box placement under simulator. During the 3 wk intervals between the first six simulations, the soil was watered with pH-adjusted water to sa turation and the watertable was allowed to drop to the bottom of the boxes before watering again. A seventh simulation was run to evaluate the effect of maintaining the soil in a saturated condition during a 4 wk period after the sixth simulation was run. After the sixth simulati on, vegetation was cut to the soil su rface and the watertable was raised to cover the surface. Water was added as needed over 4 wk to maintain soil saturation. A plastic cover was added to reduce evaporation from the boxes. The watertable was allowed to subsid e approximately 5 cm during the 2 d prior to

PAGE 37

26 the 7th simulation to facilitate moving the boxes to the simulator. The variation to the study was made as a result of recent research at the MacArthur Agro-ecology Research Center indicating that longer saturation tim es may elevate SP losses (Res. Comm., P.J. Bohlen, MacArthur Agro-ecology Research Center). Experimental Design Five treatments, with four replications each were used (Fig. 3-4). Soil was placed in the simulation boxes either as discrete laye rs (treatments C1, T1, and T2) or with the two depths mixed (treatments C2 and T3). The Al-WTR was applied on the soil surface (T1), mixed with the 0-10 cm soil depth (T2), or mixed with the combined soil depths to a depth of 20 cm (T3). Figure 3-4. Treatment descriptions No Al-WTR applied. C2** Al-WTR incorporated into 0-20 cm soil depth. T3** Al-WTR incorporated into 0-10 cm soil depth. T2* Al-WTR surface applied. T1* No Al-WTR applied. C1* *0-10 and 10-20 cm soil depths placed in box in sequence. **0-10 and 10-20 cm soil depths mixed prior to placement in box. 10-20 cm 0-10 cm Al-WTR 0-20 cm

PAGE 38

27 Soil Sampling and Forage Harvest A soil core from each packed box from both depths (0–10 and 10–20 cm) was taken prior to the start of rainfall simulations (refilled based on soil depth) and after the sixth (5 boxes) or seventh rainfall simulati ons (16 boxes). Before each rainfall simulation, the forage was harvested to a height of appr oximately 15 cm. The cuttings were dried, weighed, ground to 1 mm, and stored. Upon co mpletion of the sixth simulation, the roots were harvested, dried, weighed, and ground from replicate 4. After the seventh simulation, all other roots were harvested, dried, weighed, and gr ound. All vegetation (initial sprigs, grass cuttings, and roots) were analyzed for total P, N, and Al. Analytical Methods Soil The soil samples were dried and analyzed for WEP, TP, Mehlich-1 P, and pH. Water extractable P in the soil was determined on air dried soil. Three grams of soil were weighed into 50 mL centrifuge tubes and 30 mL of DDI water was added to obtain a soil to water ratio of 1:10. The suspension wa s then continuously shaken on a mechanical shaker for 1 hr and vacuum filtered (0.45 m ). Analysis was identical to the water samples. Soil TP was determined by the igniti on method (Anderson, 1976). One gram of soil was weighed into a 50 mL beaker and ashed in a muffle furnace. The furnace temperature was raised to 250 C and maintained for 30 min before ramping to 550C for 4 hr. After ashing, the samples were brought to room temperature in a desiccator and weighed. The cooled ash was moistened w ith distilled dionized water (DDI) before adding and evaporating 20 mL of 6.0 M HC l slowly on a hotplate at approximately 120C. Once digested, the ash was reso lublized with 2.25 mL of 6.0 M HCl and

PAGE 39

28 quantitatively transferred a nd filtered into 50 mL volumetric flasks using Whatman #41 paper. The beaker and the filter paper were each rinsed three times before bringing the flask to volume (Anderson, 1976). Twenty millili ter sub-samples were taken and stored in scintillation vials at ro om temperature. The Murphy-Riley method was used to determine TP in solution using a Technicon TM Autoanalyzer, EPA Method 365.1 (EPA, 1993a). Mehlich-1 extractable P, Fe, Al, Ca, a nd Mg were determined on air dried soil samples from both depths (0–10 cm and 10–20 cm) of each box prior to simulations and after simulation 7. Four grams of soil were weighed into 50 mL centrifuge tubes and 16 mL of Mehlich-1 solutio n (0.025 M HCl and 0.0125 M H2SO4) was added and shaken continuously on a mechanical shaker for 5 min (Mehlich, 1953). The samples were then vacuum filtered through a 0.45 m fi lter and analyzed on a TechniconTM Autoanalyzer. The pH of the soil was measured on the supernat ant of a 1:2 soil to so lution ratio (DDI). The samples were stirred and allowed to equi librate for 30 min before determining the pH using an Orion pH electrode. Aluminum Water Treatment Residuals Total P (TP) in the Al-WTR was determined via the ignition (ashing) method (Anderson, 1976) and analyzed for P followi ng the Murphy and Riley (1962) method. The Al-WTR samples were digested follo wing the EPA method 3050A and analyzed for Al, Fe, Ca, and Mg by inductively coupled plasma atomic emission spectroscopy (ICP) (USEPA, 1993b). Standard methods were used to determine pH (1:2 soil to solution) and percent solids (Page, 1982). The pH of the Al-WTR samples was meas ured on the supernatant 1:2 soil to solution ratio (DDI). The samples were st irred and allowed to equilibrate for 30 min

PAGE 40

29 before the pH was taken by the Orion pH elect rode. Percent solids were determined for the Al-WTR by taking the oven drie d weight over the air dried we ight. This is a critical factor in determining the air dried application rate of the Al-WTR. In addition, percent solids were determined for all water samp les during the TP procedure by taking the evaporated beaker weight over the know n volume, assuming 1 mL equals 1 g. Forage All forage samples were digested using a modification of the standard Kjeldahl procedure at the Forage Evaluation Support La boratory (FESL) at the University of Florida. Ground samples (0.25 g) were weighe d into 75 mL digesti on tubes and 1.5 g of the 9:1 catalyst (K2SO4:CuSO4) was added. Using 4.5 mL of H2SO4 and 2 mL of H2O2, the forage was digested for at least 4 hr at 375 C (Gallaher et al., 1975). Upon completion of the digestion, the samples were brought to volume and filtered using Fisher screening column paper. A 20 mL sub-sample taken and was stored at room temperature until it could be analyzed for total N (TN) by semiautomated colorimetry (Hambleton, 1977). Runoff, Subsurface Flow, Leachate All water samples (runoff, subsurface flow, and leachate) were vacuum filtered through a 0.45 m filter within 5 hr of collection and stored in 20 mL scintillation vials at 4 C. Soluble P (SP) was analyzed using the filtered runoff samples. Total P (TP) by digestion (see ‘Soils’) and pH analysis were performed on the unfiltered sample. In addition, the runoff, subsurface flow, and l eachate were analyzed for total Al using unfiltered samples. The data were used to de termine particulate P ( PP) in the runoff by subtracting soluble P (SP) from TP (TPSP=PP). Water soluble P was determined by

PAGE 41

30 EPA Method 365.1 using a Technicon TM Autoanalyzer, (USEPA, 1993a). Analyses was performed within 48 hr of sample collection. Water total P (TP) was determined based on a variation of the ignition (ashing) method (Anderson, 1976). Approximately 20 to 30 mL of water wa s poured into a 50 mL beaker and evaporated at 100C. The sa mples were then placed in a muffle furnace and digested and sub-sampled identical to the soil samples. The Murphy-Riley method was used to determine TP in subsample solution using a Technicon TM Autoanalyzer (EPA Method 365.1, 1993a). Total Al concentration of the water samples was determined by a nitric acid digestion procedure (Clesceri et al., 1989). One hundred millilite rs of runoff or 50 mL of subsurface and leachate were measured into a 125 mL erlenmeyer flask with two or three glass beads. Ten milliliters of 12 M HNO3 acid were added to the flask. The solution was evaporated on a hot plate until approxima tely 10 to 20 mL remained. The flask was then brought up to approximately 75 mL of so lution using DDI water and 10 mL of 12 M HNO3 acid was added again and allowed to eva porate to 10 to 20 mL of solution. The solution was then filtered using Whatman #42 pa per. Samples were then stored at room temperature until analyzed for Al by atomic absorption spectroscopy. All water samples were stored at 4C until the pH could be measured using an Orion pH electrode (Orion Research Inc. Boston, MA), usually within a week of collection. The duplicate filtered water samples in 20 mL scinti llation vials were acidified using H2SO4 and stored at 4 C until analysis fo r ammonium and nitrate. Ammonium was analyzed on a TechniconTM Autoanalyzer following EPA method 350.1 (USEPA,

PAGE 42

31 1993c). Nitrate was analyzed on an Alpke m Corp. Rapid Flow Analyzer following method A303-S170 (Alpkem, 1990). Quality Assurance and Quality Control Quality assurance and quality control (QA/QC) were met though 10% repeats, spikes, and blanks for each procedure. Cert ified external standards were used for quality control, in addition to, standa rd calibration curves. A 10% relative standard deviation was required for all repeats. Less than 5% of the total samples required re-runs because few were out-of-range values. All spike valu es and quality control checks fell within 90110% acceptance levels. Statistical Analysis Statistical analysis was performed using SAS version 8.2 1999–2001 (SAS Institute Inc., Cary, North Carolina, USA). A repeated measures model was explored to estimate and examine time (simulation) effects as well as expected correlations between treatments and their respective control for a given simulation. Normality was checked, and log-transformation was performed on th e data (variables) to achieve normal distribution. The General Line ar Model procedure was used in order to analyze the changes (variables) over time (simulation), between treatments, and between type. The means values for (variables) were compared by treatment/type/simulation using the Waller-Duncan procedure. A paired-t test was conducted in order to compare the “pre” and “post” values for the response variables measured.

PAGE 43

32 CHAPTER 4 RESULTS AND DISCUSSION Initial Soil P Concentrations and pH The initial soil was highly impacted with animal manure as indicated by the high concentrations of water extractable P (WEP), Mehlich-1 extractable P, and total P (TP) (Table 4-1). Mehlich-1 extractabl e P concentrations above 60 mg kg-1 would require no further addition of P from an agronomic standpoint. An un-impacted Immokalee fine sand (A horizon) would like ly have WEP concentrations of less than 1 mg kg-1 and Mehlich-1 extractable P in the range of 3 to 8 mg kg-1 (Graetz and Nair, 1995). Water extractable P concentrations and pH were similar between the 0–10 and 10–20 cm depths; however, Mehlich-1 P and TP concentr ations were greater in the 10–20 cm depth than in the 0–10 cm depth. Table 4-1. Initial soil mean values of P c oncentrations and pH of the Immokalee fine sand used in the study. Depth (cm) WEP (mg kg-1) Mehlich-1 P (mg kg-1) TP (mg kg-1) pH 0–10 18 879 1301 6.7 10–20 17 1034 1427 7.0 Aluminum Water Treatment Re sidual Characteristics The applied Al-WTR had a moisture cont ent of approximately 30%, a pH of 5.9, and was dominated by Al (51,000 mg Al kg-1) (Table 4-2). Miyittah-Kporgbe (2004) reported approximately 80% of the Al wa s amorphous in a similar material. Concentrations of Fe, Ca, and Mg were lo w compared to Al (Table 4-2). The TP concentration was approximately 1300 mg kg-1. This relatively hi gh TP concentration is

PAGE 44

33 not expected to affect the P dynamics in the soil because of its low availability. Makris (2004) reported desorption of P by Al-WTR was less than 1%. A Hillsborough Al-WTR as reported by O’Connor et al. (2002) ha d a low degree of P saturation (DPS: 0.032), which suggests active Fe a nd Al for P retention. Table 4-2. Selected characteristi cs of the Manatee County Al-WTR. % Solids pH Fe mg kg-1 Al mg kg-1 Ca mg kg-1 Mg mg kg-1 TP mg kg-1 WEP mg kg-1 71 5.9 1,790 51,000 580 120 1,300 BDL† † Below detectable limit: 0.02 mg L-1. Runoff, Subsurface Flow, and Leachate Characteristics Volume Runoff, subsurface flow, and leachate volum es were measured for each simulation event to provide the basis for loading calcula tions (Table 4-3). Runoff volumes averaged 13.9 L per event with no differences in runo ff volumes between treatments. Subsurface flow volumes were considerably smaller th an runoff volumes (1.2 L average) and were uniform as well, except for treatment C2 which was greater than the other treatments. The higher volume in C2 was attributed to slight differences in box construction1 and soil packing, which resulted in slightly more st anding water remaining in the C2 treatment boxes after runoff ceased. Leachate from th e bottom ports was collected only for simulations 5 through 7. Leachate volume wa s smaller than subsurface flow volume but was uniform between treatments. The uniformity of the runoff, subsurface flow, and leachate indicate the Al-WTR does not signifi cantly influence the soil’s water holding capacity at the rate applied. 1 Treatment C2 was added to the study after boxes were already constructed for all other treatments. The boxes for treatment C2 were slightly taller than the ot her boxes, resulting in an increase of water held on the soil surface after runoff ceased. This was reflected in increased subsurface flow.

PAGE 45

34 Table 4-3. Average runoff, subsurface flow and leachate volumes as influenced by treatment from six simulations of runoff, five simulations of subsurface flow, and two simulations of leachate. Treatments† C1 T1 T2 T3 C2 Average Collection Point --------------------------------L---------------------------------Runoff 13.97a‡ 13.95a 13.76a 13.88a 13.79a 13.87 Subsurface Flow 1.18b 1.13b 1.12b 1.02b 1.54a 1.20 Leachate 0.40b 0.54b 0.39b 0.63a 0.63a 0.52 †C1, no Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-WTR mixed in 0–10 cm; T3, Al-WTR mixed in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. ‡ Means within a row and followed by the same letter are not significantly different by Waller-Duncan ( p<0.05 ). pH Runoff pH decreased from 6.76 for the fi rst simulation to 6.02 in the sixth simulation (Fig. 4-1A). In contrast, subs urface flow pH increased from 7.05 to 7.66 during the same period. Leachate pH, measured for the final two simulations, was slightly greater, but similar to, subsurface flow pH. Averaged over the six simulations, there were no differences in pH among treatment s for any of the measured components ( p<0.05 ; Fig. 4-1B). Results showed pH was significantly different between collection points during each simulation and averaged over all simulations. Phosphorus Soluble P concentrations in surface r unoff were smaller than corresponding concentrations in subsurface flow and leachate for all treatments (Table 4-4). Soluble P concentrations in the subsurface flow and leach ate were at least 10-fo ld greater than in the runoff. This can be attributed to gr eater volumes of rainfall appearing as runoff (diluting the SP) than appearing as subsurface fl ow and leachate and le ss interaction with the P-impacted soil. Runoff SP concentra tion was reduced by 77% by the addition of AlWTR to the soil surface (T1). An appr oximately 45% reduction in runoff SP

PAGE 46

35 concentration was observed when the Al-WTR was mixed with the 0–10 cm soil depth (T2) and 0–20 cm soil depth (T3). 5 5.5 6 6.5 7 7.5 8 8.5 123456 SimulationpH Runoff Subsurface Leachate 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 C1T1T2T3C2 TreatmentpH Runoff Subsurface Flow Leachate Figure 4-1. Runoff, subsurface flow, and l eachate pH values from six, five, and two simulations, respectively. (A) Average treatment pH per. (B) Simulation averages for each treatment. C1, no Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-WTR mixe d in 0–10 cm; T3, Al-WTR mixed in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. Er ror bars represent the standard deviation of 4 replicates. Mean pH value within a treatment followed by the same letter are not significantly diffe rent using the Waller-Duncan method ( p<0.05 ). a a a a a (A) (B) b b b b b c c c c c

PAGE 47

36 Surface application of Al-WTR (T1) did not reduce SP concentrations in subsurface drainage or leachate (Table 4-4) Phosphorus concentrations for subsurface flow were reduced 37% by mixing the Al-WTR in the top 10 cm of soil (T2) and 90% by mixing the Al-WTR with the whole soil depth (T3). This difference between T2 and T3 was unexpected because the subsurface flow wa s collected at a depth of 10 cm and it was anticipated that SP concentrations in the s ubsurface flow would be influenced primarily by SP concentration reductions in the 0-10 cm depth. However, it appears that the SP concentration of the subsurface flow was influenced by both the 0-10 and 10-20 cm soil depths. This may be explained by the fact that the watertable was moving upward through the 10-20 cm depth during part of the simulation event allowi ng water from the 0-10 cm depth containing Al-WTR to mix with water from the untreated 10-20 cm depth. The SP concentration in subsurface flow and l eachate in T1 increased slightly (11%) for some unexplained reason. Treatments C1, T1, and T2 had similar leachate SP concentrations, whereas C2 and T3 SP concen trations were significa ntly different. AlWTR treatment of the complete profile (T 3) had the greatest effect on subsurface drainage and leachate SP concentrations with reductions of 90 and 95%, respectively. Table 4-4. Average soluble P concentrations by treatment in runoff (six events), subsurface flow (five events), and leachate (two events). Treatments† C1 T1 T2 T3 C2 Collection Point ----------------------------mg P L-1---------------------------Runoff 0.34a 0.08d 0.18c 0.16c 0.29b Subsurface Flow 4.13b 4.57a 2.60c 0.41d 4.16b Leachate 4.52b 4.63b 4.02b 0.36c 6.85a † C1, no Al-WTR, not mixed; T1, surface applied Al -WTR, not mixed; T2, Al-WTR mixed in 0–10 cm; T3, Al-WTR mixed in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. ‡Means within a row and followed by the same lette r are not significantly different by Waller-Duncan ( p<0.05 ).

PAGE 48

37 Soluble P concentrations in runoff, subsurface flow, and leachate for each simulation event are shown in Fig. 4-2 A, B, and C, respectively. Concentrations were greatest and most variable in the first two simulations in runoff, subsurface flow, and leachate. These greater concentrations may be attributed to the presence of highly-labile soil P that interacted with the rainfall duri ng the first two simulati on events. In addition, disturbance of the soil during sampling a nd repacking of the simulator boxes likely accounted for the variability obs erved during the first two simu lations. After the first two simulation events, concentrations of all treatments remained relatively stable for surface runoff, subsurface flow. This consistency reinforces results from treatment averages described previously. Runoff SP concentrations (Figure 4-2A) were the greatest in the first two simulations when compared to the following f our simulations. In simulations 3 through 6, runoff SP concentrations in all treatments were stable and did not significantly differ between treatments. After simulation 2, the control SP concentrations were not significantly different and stabi lized at approximately 0.3 mg L-1. When Al-WTR was mixed with soil (T2 and T3), runoff SP concentr ation stabilized after two simulations at approximately 0.15 mg L-1. When Al-WTR was surface-ap plied (T1), SP concentration was less than 0.1 mg L-1 and was significantly lower than all other treatments for all simulations.

PAGE 49

38 0.0 0.1 0.2 0.3 0.4 0.5 0.6 123456 SimulationSP, mg L-1 C1 T1 T2 T3 C2 0 1 2 3 4 5 6 7 23456SimulationSP, mg L-1 C1 T1 T2 T3 C2 0 2 4 6 8 10 12 56SimulationSP, mg L-1 C1 T1 T2 T3 C2 Figure 4-2. Average treatment SP concentrations and standard deviations from the three collection points. (A) Runoff: six si mulations. (B) Subsurface flow: five simulations. (C) Leachate: two simula tions. C1, no Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T 2, Al-WTR mixed in 0–10 cm; T3, AlWTR mixed in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. Similar stabilization occurred with SP c oncentrations in subsurface flow after simulation 2 (Figure 4-2B) for all treatments. Soluble P concentrations stabilized at approximately 4 mg L-1 after simulation 2 for the surface-applied Al-WTR (T1) and for the controls (C1 and C2). When the Al-WTR was incorporated in to the soil (T2 and T3), (B) (C) (A)

PAGE 50

39 SP concentrations stabili zed at 2.5 and 0.4 mg L-1, respectively. The decision to collect leachate after simulation 4 limits interpretation of the leachate data. However, with the exception of C2, the leachate SP concentratio ns were relatively consistent between simulations 5 and 6. Soluble P concentrations for C2 were expected to be similar to C1, but instead were higher. The reason for this is not obvious, but this difference might be attributed to the higher water extractable P (WEP) and Mehlic h-1 concentrations found in C2 (Tables 4-11 and 4-12). After simulation six, the soil was maintain ed in a flooded condition for 4 wk prior to simulation seven. Contrary to expect ations, flooding the soil reduced runoff SP concentration by approximately 65% (Table 45) compared to the first six simulations (Table 4-4). This effect was believed to be a result of longer interaction time with soil/Al-WTR sorption sites. Starting the rainfall event with the soil flooded likely minimized the rainfall interaction with th e soil solution which resulted in lower P concentrations in the runoff. Flooding of the soil also slightly reduced subsurface SP concentrations compared to the previous six simulations. No impact of flooding on SP concentrations was observed for the leachate. Table 4-5. Average soluble P concentrations in runoff, subsurface flow, and leachate after 4 wk of flooded conditions (Simulation 7). Treatments† C1 T1 T2 T3 C2 Collection Point ----------------------------mg P L-1---------------------------Runoff 0.13a‡ 0.04b 0.04b 0.03b 0.15a Subsurface 3.53a 3.43a 1.61b 0.20c 2.69a Leachate 3.96b 4.20b 3.49bc 0.23c 7.95a † C1, no Al-WTR, not mixed; T1, surface applied Al -WTR, not mixed; T2, Al-WTR mixed in 0–10 cm; T3, Al-WTR mixed in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. ‡Means within a row and followed by the same lette r are not significantly different by Waller-Duncan ( p<0.05 ).

PAGE 51

40 Runoff, subsurface flow, and leachate volum es for simulation 7 followed the same trend of consistency between treatments as was observed for the first 6 simulations (Table 4-6A). Flooding the soil lowered th e average runoff pH from 6.8 (simulations 16) to 6.3 (simulation 7) (Table 4-6B). Aver age subsurface flow and leachate pH values decreased from approximately 7.6 to 7.3. Table 4-6. Runoff, subsurface flow, and l eachate collected during simulation 7. (A) Volume. (B) pH. (A) Treatments† C1 T1 T2 T3 C2 Collection Point ---------------------------------L--------------------------------Runoff 15.81b‡ 16.78b 15.23c 15.83b 16.63b Subsurface 0.77a 0.86a 1.07b 1.11ab 1.22b Leachate 0.42a 0.73a 0.36a 0.63a 0.57a (B) Treatments† C1 T1 T2 T3 C2 Collection Point -------------------------------pH-------------------------------Runoff 6.34a‡ 6.10a 6.44a 6.30a 6.38a Subsurface 7.37b 7.42b 7.39b 7.31b 7.19b Leachate 7.44c 7.44c 7.59c 7.35c 7.34c † C1, no Al-WTR, not mixed; T1, surface applied Al-WTR, not mixed; T2, Al-WTR mixed in 0–10 cm; T3, AlWTR mixed in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. ‡Means within a row and followed by the same letter are not significantly different by Waller-Duncan ( p<0.05 ). Soluble P loss per box from each treatment per rainfall event was calculated by multiplying volume by concentration (Table 4-7). The same trend was observed as with SP concentrations, which was expected base d on volume similarities among treatments (Table 4-6A). These results emphasize the importance of treating P-impacted soils to reduce risk of P loss. Mixi ng 2.5% Al-WTR with the P-im pacted soil can reduce field loss of SP by approximately 80% compared to 30% by surface applie d Al-WTR. Surface application and partial mixing of the impacted zone (T1 a nd T2, respectively) had less impact compared to total mixing of Al-W TR with the P-impacted soil (T3). The relatively lower volumes of subsurface and leach ate pose an equal risk of SP loss to that

PAGE 52

41 of runoff because what is l acking in volume is compensated by concentration. A rough estimate of SP loss on a hectare basis can be calculated, assuming that similar P losses would be encountered over larg er areas (Table 4-7). The ex pected loss per rainfall event in an untreated field would be 350 to 500 g ha-1. Partial treatment of the impacted zone could reduce P loss approximately 200 to 300 g ha-1. Soluble P loss could be reduced to less than 100 g ha-1 if Al-WTR was incorporated th roughout the P-impacted layer. Soil mixing in C2 did not reduce P loss as was found by Miyittah (2004), because both the 0–10 and 10–20 cm depths were highly P-impacted. Miyittah (2004) mixed a Pimpacted soil with an un-impacted E horizon, wh ich in effect, diluted the P-impacted soil. Table 4-7. Soluble P mass loss (SP concentration volume) av eraged over six rainfall events and estimated SP loss pe r hectare per rainfall event. Treatments† Collection Point C1 T1 T2 T3 C2 Runoff (mg P box-1) 4.36a‡ 1.07d 2.22c 1.96c 3.68b Subsurface (mg P box-1) 4.75b 4.92b 2.76c 0.45d 6.07a Leachate (mg P box-1) 1.69b 2.37b 1.50b 0.20c 4.61a Total SP loss (mg P box-1) 10.80 8.36 6.48 2.61 14.36 SP loss (g P ha-1) 360 279 216 87 479 † C1, no Al-WTR, not mixed; T1, surface applied Al -WTR, not mixed; T2, Al-WTR mixed in 0–10 cm; T3, Al-WTR mixed in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. ‡Means within a row and followed by the same lette r are not significantly different by Waller-Duncan ( p<0.05 ). Particulate P (PP) concentration was calc ulated in runoff, subsurface flow, and leachate by subtracting SP from TP (Table 4-8). Runoff PP concentrations ranged from 0.1 to 0.3 mg L-1. On average, PP accounted for appr oximately 20 and 60% of TP in the controls and the Al-WTR treatments, resp ectively. Therefore, PP should also be a concern on minimally sloped landscapes, par ticularly if Al-WTR has been applied. Particulate P concentrations in the subsurf ace flow and leachate were below detectable limits. Although PP concentrations in runoff showed marked trends among treatments, the differences were not significant because of the high standard deviations among

PAGE 53

42 treatments. This is believed to be a resu lt of small variations in soil surface levels compared to the box edge rather than treatment effects. Th is would affect the amount of particulate matter leaving, or conversely, retained, at th e box edge where runoff was collected. Table 4-8. Total, soluble, a nd particulate P concentrations in runoff for each treatment averaged over six rainfall events. Treatments† C1 T1 T2 T3 C2 ------------------------------mg P L-1-----------------------------Total P 0.47a‡ 0.21b 0.38ab 0.47a 0.34ab Soluble P 0.34a 0.08d 0.18c 0.16c 0.29b Particulate P 0.13a 0.13a 0.20a 0.31a 0.05a † C1, no Al-WTR, not mixed; T1, surface applied Al -WTR, not mixed; T2, Al-WTR mixed in 0–10 cm; T3, Al-WTR mixed in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. ‡Means within a row and followed by the same lette r are not significantly different by Waller-Duncan ( p<0.05 ). Nitrateand Ammonium-Nitrogen Ammonium-N concentrations in runoff, subsurface flow, and leachate averaged less than 1 mg L-1 and there were no differences among treatments (Table 4-9A). NitrateN concentrations (Table 4-9B ) were also below 1 mg L-1 and showed no differences between treatments with two excepti ons. The exceptions were greater NO3-N concentrations in subsurface flow and leachate in T1 compared to all other treatments. There are no obvious explanations for these differences at this time. Under typical production conditions, N fertilizer would be added to maintain production and forage nutritive value. However, to simplify this experiment, no additional fertilization was used.

PAGE 54

43 Table 4-9. Average N concentrations in runo ff, subsurface flow, and leachate from six, five, and two simulations, respectively. (A) NH4-N. (B) NO3-N. (A) Treatments† C1† T1 T2 T3 C2 Collection Point ------------------------mg NH4-N L-1-----------------------Runoff 0.19a‡ 0.19a 0.18a 0.23a 0.19a Subsurface 0.38a 0.38a 0.35a 0.30a 0.33a Leachate 0.35a 0.36a 0.42a 0.47a 0.44a (B) Treatments† C1† T1 T2 T3 C2 Collection Point ------------------------mg NO3-N L-1------------------------Runoff 0.15a‡ 0.14a 0.14a 0.13a 0.12a Subsurface 0.23b 0.77a 0.10b 0.14b 0.13b Leachate 0.12b 0.74a 0.11b 0.10b 0.08b † C1, no Al-WTR, not mixed; T1, surface applied Al -WTR, not mixed; T2, Al-WTR mixed in 0–10 cm; T3, Al-WTR mixed in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. ‡Means within a row and followed by the same lette r are not significantly different by Waller-Duncan ( p<0.05 ). Total Calcium, Magnesium, Iron, and Aluminum Concentrations Calcium and Mg concentrations were in the order of runoff < subsurface flow < leachate (Table 4-10). There were no significant differences in Ca and Mg concentrations in runoff among treatments. Calcium and Mg concentrations were different among treatments for subsurf ace flow and leachate. Aluminum-WTR treatments had higher concentrations of Ca in subsurface flow and leachate which is likely due to the Ca containe d in the Al-WTR (580 mg L-1). Iron concentrations were similar in runoff, subsurface flow, and leachat e, and did not vary between treatments. Total Al concentrations were highly variab le and showed no meaningful differences among treatments. However, a trend of hi gher Al concentrations was observed in leachate from treatments where Al-WTR was in corporated into the soil (T2 and T3). Random samples were tested for soluble Al from all collection points, treatments, and simulations and all concentrations we re below detection limits (0.3 mg L-1). This suggests that essentially all of the Al was in particulate form.

PAGE 55

44 Table 4-10. Average total Ca, Mg, Fe, and Al concentrations in runoff, subsurface flow, and leachate from six, five, and two simu lations, respectively. (A) Runoff. (B) Subsurface flow. (C) Leachate. (A) Runoff Ca Mg Fe Al Treatments† -------------------------mg L-1------------------------C1 15.8 8.83 0.27 0.66 T1 16.3 8.90 0.53 1.06 T2 16.4 8.44 0.43 1.62 T3 17.6 8.68 0.51 2.68 C2 15.7 8.36 0.23 2.71 (B) Subsurface Flow Ca Mg Fe Al Treatments† ------------------------mg L-1-----------------------C1 64.3c‡ 22.1b 0.23 0.00b T1 80.2b 26.8ab 0.42 1.22a T2 91.1ab 30.6a 0.34 0.00b T3 95.9a 31.6a 0.38 0.00b C2 49.0d 16.4c 0.54 0.76a (C) Leachate Ca Mg Fe Al Treatments† ------------------------mg L-1---------------------C1 95.2b‡ 45.8b 0.19 0.00a T1 94.4b 48.7b 0.09 0.00a T2 138.8ab 67.9a 0.15 1.94b T3 199.3a 63.0a 1.14 5.67b C2 87.7b 28.1c 0.12 3.70b † C1, no Al-WTR, not mixed; T1, surface applied Al -WTR, not mixed; T2, Al-WTR mixed in 0–10 cm; T3, Al-WTR mixed in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. ‡Means within a column and followed by the same letter are not significantly different by Waller-Duncan ( p<0.05 ). Preand Post-simulation Comparisons Soil Phosphorus When Al-WTR was incorporated into the soil, WEP decreased by approximately 70% i.e., T2 (0–10 cm depth) and both depths in T3, compared to th e respective controls (Table 4-11). There was no change in soil WE P concentrations in any of the treatments prior to or after the study (paired T Test; p< 0.05).

PAGE 56

45 Table 4-11. Preand post-simu lation water extractable P conc entrations in the 0-10 and 10-20 cm soil depths. 0–10 cm depth 10–20 cm depth Treatments† Pre-simulation‡§ Post-simulation Pre-simulation Post-simulation -----------------------------mg P kg-1----------------------------C1 21 a 20 a 24 a 26 a T1 14 c 14 b 24 a 23 ab T2 7 d 7 c 19 b 18 b T3 6 d 7 c 7 c 6 c C2 25 a 24 a 24 a 25 a † C1, no Al-WTR, not mixed; T1, surface applied Al -WTR, not mixed; T2, Al-WTR mixed in 0–10 cm; T3, Al-WTR mixed in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. §Means within a column followed by the same lette r are not significantly different by Waller-Duncan ( p<0.05 ). Application of Al-WTR had essentially no effect on Mehl ich-1 extractable P (Table 4-12). Apparently, Mehlich-1 extractant remove s P adsorbed to the soil as well as that adsorbed to Al-WTR. There were few differences in preand post-simulation concentrations statistically compared using the paired T Test ( p<0.05 ). Table 4-12. Preand post-simu lation Mehlich-1 P concentra tions in the 0-10 and 10-20 cm soil depths. 0–10 cm depth 10–20 cm depth Treatments† Pre-simulation‡§ Post-simulation Pre-simulation Post-simulation -----------------------------mg P kg-1----------------------------C1 679 AB 785 B 882 A 882 BC T1 635 AB 774 B 825 AB 1017 A T2 614 B 709 B 760b AB 954a AB T3 779 A 770 B 720 B 795 C C2 777a A 958b A 823 AB 903 B † C1, no Al-WTR, not mixed; T1, surface applied Al -WTR, not mixed; T2, Al-WTR mixed in 0–10 cm; T3, Al-WTR mixed in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. ‡ Means for each soil depth within a row followed by the same lower case letter are not significantly different by Waller-Duncan ( p<0.05 ). § Means for each soil depth within a column followed by the same upper case letter are not significantly different by Waller-Duncan ( p<0.05 ). Soil pH Soil pH decreased slightly by the end of the study (Table 4-13). However, application of Al-WTR had e ssentially no effect on soil pH before or after simulation

PAGE 57

46 events. It appears the soil was well-buffered due to the Ca and Mg accumulated from the manure applications. Table 4-13. Preand post-simulation soil pH values in the 0-10 and 10-20 cm soil depths. 0–10 cm depth 10–20 cm depth Treatments† Pre-simulation‡§ Post-simulation Pre-simulation Post-simulation C1 6.9a A 6.6b A 7.0a AB 6.7b A T1 6.7 AB 6.6 A 7.0a A 6.6b A T2 6.9 B 6.6 A 6.9a ABC 6.6b A T3 7.0a A 6.6b A 6.8 C 6.6 A C2 6.9a A 6.6b A 6.8a BC 6.6b A † C1, no Al-WTR, not mixed; T1, surface applied Al -WTR, not mixed; T2, Al-WTR mixed in 0–10 cm; T3, Al-WTR mixed in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. ‡ Means for each soil depth within a row followed by the same lower case letter are not significantly different by Waller-Duncan ( p<0.05 ). § Means for each soil depth within a column followed by the same upper case letter are not significantly different by Waller-Duncan ( p<0.05 ). Forage Shoot yields were smaller for the surface-applied Al-WTR (T1) than for incorporated Al-WTR (T2 and T3) treatments fo r the initial three harv ests (Table 4-14). This might suggest that surface-applied Al-W TR was detrimental to stargrass growth. However, this same trend occurred for control C2. This suggests that the differences may be due to initial stand establishment rather th an toxicity effects. After the first two simulations (63 d of growth), the stargrass became well-established. For simulations 4-6, shoot yields were similar between treatments There was no visual evidence of adverse effects on stargrass shoot gr owth. In addition, root mass of all treatments was approximately 30 g box-1 during the experiment (Figure 4-3). Root biomass was not significantly different among treat ments; therefore Al-WTR doe s not appear to adversely affect stargrass roots.

PAGE 58

47 Table 4-14. Stargrass shoot biomass harvested prior to each simulation. Treatments† Sim 1 Sim 2 Sim 3 Sim 4 Sim 5 Sim 6‡ Avg -----------------------------------g box-1----------------------------------C1 35a§ 17a 38a 16a 8a 57a 23 T1 17b 8b 24b 19a 4b 41a 14 T2 35a 21a 35ab 18a 7a 50a 23 T3 31a 14a 30ab 17a 7a 56a 20 C2 14b 6b 28ab 21a 9a 47a 15 † C1, no Al-WTR, not mixed; T1, surface applied Al -WTR, not mixed; T2, Al-WTR mixed in 0–10 cm; T3, Al-WTR mixed in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. ‡Simulation 6 data includes all above-ground biomass. Simulation 1–5 data are shoot biomass above 15 cm height. §Shoot biomass was compared (column) between treatments using a log transformation followed by Waller-Duncan method. Treatments with the same letter are not significantly different ( p<0.05 ). 0 20 40 60 80 100 120 140 C1T1T2T3C2TreatmentGrams Box-1 Shoots Roots Figure 4-3. Total shoot and root growth per treatment averaged over six simulations. C1, no Al-WTR, not mixed; T1, surface a pplied Al-WTR, not mixed; T2, AlWTR mixed in 0–10 cm; T3, Al-WTR mixed in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. Error bars represent the standard deviation of 4 replicates. Treatments means within a plant fraction and followed by the same letter are not significantly different us ing the Waller-Duncan method ( p<0.05 ). Shoot and root P concentrations among all treatments averaged over the six simulations were not different ( p<0.05 ; Figure 4-4A). During the first two simulations, the stargrass contained a pproximately 3 mg of P g-1 of forage and declined by approximately 0.5 mg uniformly across all treatments by the end of the experiment (Figure 4-4B). The original planting material contained approximately 3.2 mg g-1. The decline was expected based on seasonal ch ange in growth (August to January). a a a a a a a ab b b

PAGE 59

48 Simulation 1 began in August and simula tion 6 ended in December. Root P concentrations were approximately 1.7 mg g-1 and did not significantly vary among treatments (Figure 4-4A). 0 1 2 3 4 C1T1T2T3C2 TreatmentsTP, mg g-1 Shoots Roots 0 1 2 3 4 123456 HarvestTP, mg g-1 Figure 4-4. Average stargrass TP concentr ations per treatment averaged over all simulations (A) and per simulation aver aged over all treatments (B). C1, no Al-WTR, not mixed; T1, surface applie d Al-WTR, not mixed; T2, Al-WTR mixed in 0–10 cm; T3, Al-WTR mixe d in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. Error bars represent the sta ndard deviation of 4 replicates. Shoot and root TN concentrations al so were not affected by treatment ( p <0.05 ; Figure 4-5A). Shoot TN concentr ation was approximately 25 mg N g-1in the planting material and for the first two simulations (Fi gure 4-5B). However, with no additional N input, the N concentration declined to approximately 10 mg N g-1. Root TN concentrations at the end of the study were similar for all treatments (Figure 4-5A). (A) (B)

PAGE 60

49 0 5 10 15 20 25 30 C1T1T2T3C2 TreatmentsTN, mg g-1 Shoots Roots 0 5 10 15 20 25 30 123456 HarvestTN, mg g-1 Figure 4-5. Average stargrass TN concentr ations per treatment averaged over all simulations (A) and per simulation aver aged over all treatments (B). C1, no Al-WTR, not mixed; T1, surface applie d Al-WTR, not mixed; T2, Al-WTR mixed in 0–10 cm; T3, Al-WTR mixe d in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. Error bars represent the st andard deviation of 4 replicates Shoot Al concentrations fluctuated slight ly between treatments and simulations, but typically remained below 40 g g-1 (Figure 4-6A, B). These c oncentrations were at the low end of the typical range (60–14,500 ppm) for grasses (Pendias and Pendias, 2001). Root Al concentrations ranged from 970 to 310 g g-1 of forage. These values also fall within the typical range for forage plants. Because of the high standard deviation, there were no significant differences among treatments ( p<0.05 ). Contamination is believed to be a factor even though roots were washed multiple times to remove soil and Al-WTR (A) (B)

PAGE 61

50 particles. No stunted root s were observed. Aluminum concentration (Figure 4-6B) exhibited a decline that was similar to P and N concentrations in simulations 3 through 5. The increase in concentration in simulation 6 is attributed to shoot contamination when cutting shoots close to th e soil surface; which is s upported by the large standard deviation. 0 200 400 600 800 1000 1200 1400 1600 C1T1T2T3C2TreatmentsTotal Al, g g-1 Shoots Roots 0 50 100 150 200 250 123456 HarvestTotal Al, g g-1 Figure 4-6. Average stargrass Al concentr ations per treatment averaged over all simulations (A) and per simulation aver aged over all treatments (B). C1, no Al-WTR, not mixed; T1, surface applie d Al-WTR, not mixed; T2, Al-WTR mixed in 0–10 cm; T3, Al-WTR mixe d in 0–20 cm; C2, no Al-WTR, mixed 0–20 cm. Error bars represent the sta ndard deviation of 4 replicates. Treatments means within the root fraction and followed by the same letter are not significantly different us ing the Waller-Duncan method ( p<0.05 ). b a ab b b (A) (B)

PAGE 62

51 CHAPTER 5 CONCLUSIONS Previous research has shown that land applying Al-WTR is effective in reducing soluble P loss from agricultural fields. Th is study was developed to provide Al-WTR management information on manure-impacted sandy soils with a fluctuating high watertable. The research evaluated differe nt methods of applying Al-WTR (surface and incorporated) as well as different incorporat ion depths, P loss over time, and potential effects on forage productivity and chemical composition. Runoff boxes were filled with a P-impacted sandy soil and the respective Al-W TR application treatments. The rate of Al-WTR application (2.5% of dry weight of soil) was selected based on MiyittahÂ’s (2004) recommendation for a practical field ap plication rate that would significantly reduce soluble P concentrations in the soil so lution. Boxes were placed under a rainfall simulator to create uniform rainfall dist ribution at known intervals and duration. Phosphorus loss was examined from the three collection points (runoff, subsurface flow, and leachate) because of potential water move ment above a restricting layer (spodic). Concerns about P deficiency and Al toxicity in stargrass were addressed throughout the duration of the study. The application of Al-WTR, whether surf ace applied or mixed within the profile, impacted SP loss. Surface applied Al-WTR wa s more effective in reducing SP in runoff than when incorporated. However, incorpor ated Al-WTR was more effective in reducing SP in subsurface flow and leachate than was surface Al-WTR application. Care must be taken to ensure complete incorporation of Al-WTR throughout the P-impacted layer, as

PAGE 63

52 Al-WTR is only effective in reducing SP concentrations when it is in contact with the impacted soil. To achieve the best results for reducing P loss in both surface runoff and subsurface flow/leachate from highly impacted soils, Al-WTR should be first mixed with the impacted soil depth to reduce subsurf ace flow/leachate P loss AND then added to the soil surface to minimize P loss in runoff. For an un-impacted area with low initial soil P concentration intended for manure applica tion, surface applicati on of Al-WTR would likely suffice to minimize P loss. Application of Al-WTR at 2.5% of soil weight did not adversely affect forage yield or quality of stargrass, based on the unifo rm values of yield and P, N, and Al concentrations between treatments. Field scal e forage studies are ne eded to validate boxscale results. Effective P control on intensively managed agricultural land is imperative. The use of Al-WTRs is not a single source solution. Ra ther, it is an effective management tool in an intricate comprehensive management pl an. This study attempted to mimic field conditions to provide realistic results. However, shorta nd long -term field studies are needed before Al-WTR application can be included in best management practices.

PAGE 64

53 LIST OF REFERENCES Alpkem Corp. 1990. Rapid Flow Anal yzer Methodology. Method A303-S170. Clackamas, OR. Anderson, D.L., O.H. Tuovinen, A. Faber, and I. Ostrokowski. 1995. Use of soil amendments to reduce soluble phosphorus in dairy soils. Ecol. Eng. 5: 229–246. Atalay, A. 2001. Variation in phosphorus sorp tion with soil particle size. Soil and Sed. Contam. 10: 317–335. Basta, N.T., and J.J. Sloan. 1999. Bioavailability of heavy metals in strongly acidic soils treated with exceptional qua lity biosolids. J. Environ. Qual. 28: 633–638. Ballard, R., and J.G.A. Fiskell. 1974. Phosphorus retention in coastalplain forest soils: Relationship to soil properties. Soil Sci. Soc. of Amer. J. 38: 250–255. Boggess, C.F., E.G. Flaig, and R.C. Fluck. 1995. Phosphorus budgets for Lake Okeechobee tributary watershe ds. Ecol. Eng. 5: 143–162. Breeuwsma, A., and S. Silva. 1992. Phosphorus fertilization and environmental effects in the Netherlands and the Po regions (Italy). Rep. 57. Agric. Res. Dep., The Winand Staring Centre for Integrated La nd, Soil, and Water Res., Wageningen, the Netherlands. Brown, E., and J.B. Sartain. 2000. Phosphor us retention in United States Golf Association (USGA) greens. Soil Crop Sci. Soc. Florida Proc. 59: 112–117. Bugsbee, G.J., and C.R. Frink. 1985. Alum sludge as a soil amendment: Effects on soil properties and plant growth. New Haven Connecticut Agric. Exp. Stn. Bull. 827. Buss, G.R., J.A. Lutz, Jr., and G.W. Hawkins. 1975. Yield response of alfalfa cultivars and clones to several pH levels in Tatum subsoil. Agron. J. 67: 331–334. Clesceri, L.S., A.E. Greenberg, and R.R. Trussell (ed.). 1989. 3030D, Preliminary digestion for metals. In: Standard Me thods for the Examination of Water and Wastewater: 17th ed. Amer. Public Health Assoc. Washington, D.C. Cornell, D.A., and G.P. Westerhoff. 1981. Management of wa ter treatment plant sludges. In: Sludge and its Ultimate Dis posal. (Ed. J.A. Borchardt). Ann Arbor Sci. Publishers. Ann Arbor, MI.

PAGE 65

54 Dayton, E.A., and N.T. Basta. 2001. Char acterization of dri nking water treatment residuals for use as a soil substitute. Water Environment Research 73: 52–57. Delhaize, E., and P.R. Ryan. 1995. Aluminum toxicity and toleran ce in plants. Plant Physiol. 107: 315–321. Elliott, H.A., and B.A. Dempsey. 1991. Agronom ic effects of land application of watertreatment sludges. J. Amer. Water Works Assoc. 83: 126–131. Elliott, H.A., G.A. O’Connor, P. Lu, and S. Brinton. 2002. Influence of water treatment residuals on phosphorus solubility and l eaching. J. Environ. Qual. 31: 1362–1369. Elliott, H.A., B.A. Dempsey, D.W. Hamilton, and J.R. Dewolfe. 1990. Land application of water treatment sludges; Impacts and management. Amer. Water Works Assoc. Res. Denver, CO. Esteves, M., O. Planchon, J.M. Lapetite, N. Silvera, and P. Cadet. 2000. The ‘emire’ large rainfall simulator: design and field testing. Earth Surf. Processes Landforms 25: 681–690. Farmer, EE. 1973. Relative detachability of so il particles by simulated rainfall. Soil Sci. Soc. of Amer Proc. 37: 629–633. Faulkner, S.P., and C.J. Richardson. 1989. P hysical and chemical characteristics of freshwater wetland soils. pp. 41–72. In: Hammer, D.A. (ed.). Constructed Wetlands for Wastewater Treatment. Lewis Publishers. Chelsea, MI. Foy, C.D. 1984. Physiological effects of hydr ogen, aluminum, and manganese toxicities in acid soils. In: So il acidity and liming, (Ed. F. Adams). pp. 57–97. Agron. Monograph. 12, 2nd ed. Amer. Soc. Agron., Crop. Sci. Soc. Amer., and Soil Soc. Amer., Madison, WI. Foy, C.D., R.L. Chaney, and M.C. White. 1978. Physiology of metal t oxicity in plants. Ann. Rev. Plant Physiol. 29: 511–566. Foy, C.D., A.L. Fleming, and G.C. Gerloff. 1972. Differential aluminum tolerance in two snapbean varieties. Agron. J. 64: 815–818. Flaig, E.G., and K.R. Reddy. 1995. Fate of phosphorus in the Lake Okeechobee watershed, FL, USA: overview and r ecommendations. Ecol. Eng. 5: 127–142. Flaig, E.G., and K.E. Havens. 1995. Hist orical trends in the Lake Okeechobee ecosystem. I. Land use and nutrient load ing. Arch. Hydrobiol. Suppl. 107: 1–24. Florida Agriculture Statistical Directory. 2004. Florida Department of Agricultural and Consumer Services. Tallahassee, FL.

PAGE 66

55 Fluck, R.C., C. Fonyo, and E. Flaig. 1992. Land-use-based phosphorus balances for Lake Okeechobee, Florida, drainage wate rsheds. Appl. Eng. Agric. 8: 813–820. Froelich, P.N. 1988. Kinetic control of dissolved phosphate in natural rivers and estuaries: A primer on the phosphate bu ffer mechanism. Limnol. Oceanogr. 33: 649–668. Frossard, E., M. Brossard, M.J. Hedle y, and A.K. Metherell. 1995. Reactions controlling the cycling of P in soils. In : Phosphorus in th e Global Environment Transfers, Cycles, and Management. (Ed. H. Tiessen). John Wiley & Sons, New York, NY. Gale, P.M., K.R. Reddy, and D.A. Graetz. 1994. Phosphorus retention by wetland soils used for treated wastewater dispos al. J. Environ. Qual. 23: 370–377. Gallaher, R.N., C.O. Weldon, and J.G. Futral 1975. An aluminum block digester for plant and soil analysis. Soil Sc i. Soc. Amer. Proc. 39: 803–806. Gallimore, L.E., N.T. Basta, D.E. Storm, M.E. Payton, R. H. Huhnke, and M.D. Smolen. 1999. Water treatment residual to redu ce nutrients in surface runoff from agricultural land. J. Environ. Qual. 28: 1474–1478. Gatewood, S.E., and P.B. Bedient. 1975. Dr ainage density in the Lake Okeechobee drainage area. Report, Division of State Planning, Florida Department of Administration. Tallahassee, FL. Graetz, D.A., and V.D. Nair. 1995. Fate of phosphorus in Florida Spodosols contaminated with cattle manure. Ecol. Eng. 5: 163–182. Hambleton, L.G. 1977. Semiautomated method for simultaneous determination of phosphorus, calium, and crude protein in animal feeds. J.A.O.A.C. 60: 845–852. Haustein, G.K., T.C. Daniel, D.M. Miller, P.A. Moore, Jr., and R.W. McNew. 2000. Aluminum-containing residuals influence high-phosphorus soils and runoff water quality. J. Environ. Qual. 29: 1954–1959. Haug, A. 1984. Molecular aspects of aluminum toxicity. CRC Crit. Rev. Plant Sci. 1: 345–373. He, Z.L., A.K. Alva, Y.C. Li D.V. Calvert, and D.J. Banks. 1999. Sorption-desorption and solution concentration of phosphorus in a fertilized sandy so il. J. Environ. Qual. 28: 1804–1810. Hoddinott, J., and C. Richter. 1987. The in fluence of aluminum on photosynthesis and translocation in french bea n. J. Plant Nutr. 10: 443–454.

PAGE 67

56 Hodges, J.R., G.K. Kirk, R.L. Shirley, F.M. Peacock, J.F. Easley, H.L. Breland, and F.G. Martin, 1967. Phosphorus fertilization of pangola grass pastures, direct and residual effects. Research Report, Range Experiment Station, Univ. of FL, Ona, FL. Holford, I.C.R., C. Hird, and R. Lawie. 1997. Effects of animal effluents on the phosphorus sorption characteristics of so ils. Aust. J. Soil Res. 35: 365–373. Howard, F. 2006. Pulling out P. Hay and Forage. March 2006: 52. Humphry, J.B., T.C. Daniel, D.R. Edward s, and A.N. Sharpley. 2002. A portable rainfall simulator for plot-scale runoff studies. Appl. Eng. Agric. 18: 199–204. Ippolito, J.A., K.A. Barbarick, D.M. Heil, J.P. Chandler, and E.F. Redente. 2003. Phosphorus retention mechanisms of a wate r treatment residual. J. Environ. Qual. 32: 1857–1864. Ippolito, J.A., K.A. Barbarick, and E.F. Redente. 2002. Comb inations of water treatment residuals and biosolids aff ect two range grasses. Co mmun. Soil Sci. Plant Anal. 33: 831–844. Jarvis, S.C., and D.J. Hatch. 1986. The effects of low concentrations of aluminum on the growth and uptake of nitrate-N by wh ite clover. Plant Soil 95: 43–55. Kinraide, T.B. 1991. Identity of the rhizot oxic aluminium species. Plant Soil. 134: 167–178. Kleinman, P.J.A., and A.N. Sharpley. 2003. Effect of broadcast manure on runoff phosphorus concentration over successive rain fall events. J. Environ. Qual. 32: 1072–1081. Kleninman, P.J.A., A.N. Sharpley, B.G. Moye r, and G.F. Elwinger. 2002. Effect of mineral and manure phosphorus sources on runoff phosphorus. J. Environ. Qual. 31: 2026–2033. Klimashe, E.L. 1970. On role of roots as de termining different tolerance of genetically related plant forms toward Al3+. Agrochimica 14: 232–241. Lindsay, W.L., P.L.G. Vlek, and S.H. Chien. 1989. Phosphate minerals. pp. 1089– 1130. In: Minerals in Soils Environments. (Ed. J.B. Dixon, and S.B. Weed). Soil Sci. Soc. Amer. J., Madison, WI. McCaffery, P.M., W.M. Hinkley, R. MacGil l, and G.D. Cherr. 1976. Report of investigations in the Kissimmee-Lake Okeechobee Watershed. FL Dept. Environ. Reg. Tech. Series, Vol. 2, No. 2, Tallahassee, FL. Mehlich, A. 1953. Determination of P, Ca, Mg, K, Na, and NH4. Soil Testing Div. Publ. 1–53. North Carolina Department of Agriculture, Raleigh.

PAGE 68

57 Meyer, L.D.,and D.L. McCune. 1958. Rainfall simulator for runoff plots. Agric. Eng. 39: 644–648. Miller, W.P. 1987. A solenoid-operated, variable intensity rainfall simulator. Soil Sci. Soc. of Amer. J. 51: 832–834. Miner, C.L. 2001. Storage and partitioning of soil phosphorus in the Orlando Easterly Wetland treatment system. M.Sci. Thesis Univ. of FL, Gain esville, FL. 326110510. Mitsch, W.J., and J.G. Gosselink. 1993. Wetlands, 2nd ed. Van Nostrand Reinhold Company, NY. Miyittah-Kporgbe, M. 2004. Phosphorus immo bilization in manureimpacted soil with aluminum-based drinking water treatment residual. M.Sci. Thesis, Univ. of FL, Gainesville, FL. 32611-0510. Moore, P.A., Jr., P.B. DeLaune, D.E. Carm an, T.C. Daniel, and A.N. Sharpley. 2000. Development of phosphorus index for past ures. pp. 158–165. In: Proc. 2000 Natl. Poultry Waste Management Symp. (Ed. J.P. Blake, and P.H. Patterson). Auburn Press, Auburn, AL. Moore, P.A. Jr., D.M. Miller, T.C. Dani el, B.R. Shreve, and D.R. Edwards. 1994. Decreasing phosphorus solubility and inhi biting ammonia volatilization in poultry litter with alum. Pou ltry Science 73: 151. Munn, J.R., and G.L. Huntington. 1976. A porta ble rainfall simulator for erodibility and infiltration measurements on rugged terrain. Soil Sci. Soc. of Amer. J. 40: 622– 624. Murphy, J., and J.P. Riley. 1962. A modi fied single solution method for the determination of phosphate in natural waters. Anal. Chem. Acta. 27: 31–36. Nair, V.D., and D.A. Graetz. 2002. Phos phosrus saturation in spodosols impacted by manure. J. Environ. Qual. 31: 1279–1285. Nair, V.D., D.A. Graetz, and K.M. Portier. 1995. Forms of phosphor us in soil profiles from dairies of south Florida. Soil Sci. Amer. J. 59: 1244–1249. 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. National Phosphorus Research Project (NPR P). 2001. Rain simulator protocols. (Accessed 11/3/06). http://www.sera17.ext.vt.edu/Docu ments/National_P_protocol.pdf.

PAGE 69

58 Neff, E.L. 1979. Why rainfall simulation? In: Proceedings of Rainfall Simulator Workshop. Tucson, AZ. USDA-SEA ARM-W-10: 3–7. Novak, J.M., and D.W. Watts. 2004. Incr easing the phosphorus sorption capacity of southeastern coastal plain soils using water treatment residuals. Soil Sci. Soc. Amer. 169: 206–214. O’Connor, G.A., H.A. Elliott, and P. Lu. 2002. Characterizing water treatment residuals phosphorus retention. Soil and Crop Sci. Soc. FL Proc. 61: 67–73. Ohki, K. 1986. Photosynthesis, chlorophyll, and transpiration responses in aluminum stressed wheat and sorghum. Crop Sci. 26: 572–575. Olila, O.G., and K.R. Reddy. 1995. Influence of pH on phosphorus retention in oxidized lake sediments. Soil Sci. Soc. of Amer. J. 59: 946–959. Osborne, G.J., J.E. Pratley, and W.P. Stewar t. 1981. The tolerance of subterranean clover (Trifolium-Subterraneum L) to aluminum and manganese. Field Crops Res. 3: 347–358. Page, A.L. 1982. Methods of Soil Analysis. Part 2. Chemical and Microbiological Properties. 2nd ed. Agron. Monogr. 9. ASA and SSSA. Madison, WI. Reddy, K.R., E.G. Flaig, and D.A. Graetz. 1996. Phosphorus st orage capacity of uplands, wetlands and streams of the Lake Okeechobee watershed, Florida. Agric. Ecosystems & Environ. 59: 203–216. Reddy, K.R., and W.H. Smith (ed). 1987. A quatic plants for water treatment and resource recovery. Magno lia Publishing, Orlando, FL. Riley, S.J., and Hancock F. 1997. A rain fall simulator for hydrologic and erosion experiments on mines, with an example of its applications at Ranger Uranium Mine, Northern Territory, Australia. Th e Australasian Institute of Mining and Metallurgy Proceedings. 1: 3–8. Romkens, M.J.M., L.F. Glenn, D.W. Nelson, and C.B. Roth. 1975. A laboratory rainfall simulator for infiltration and soil detachment studies. Soil Sci. Soc. of Amer. Proc. 39: 158–160. SAS Institute, 2001. The SAS System for Wind ows. Version 8 Release 8.2 SAS Inst. Cary, NC. Samac, D.A., and M. Tesfaye. 2003. Plant improvement for tolerance to aluminum in acid soils a review. Plant Cell, Tissue and Organ Culture 75: 189–207 Sarkunan, V., C.C. Biddapa, and S.K. Nayak. 1984. Physiology of aluminum toxicity in rice. Curr. Sci. 53: 822–824.

PAGE 70

59 Sharpley, A.N., and P. Kleinman. 2003. Effect of rainfall simulato r and plot scale on overland flow and phosphorus transport. J. Environ. Qual. 32: 2172–2179. Sharpley, A.N., and R. Menzel. 1987. The impact of soil and fe rtilizer phosphorus on the environment. Advances in Agron. 41: 297–324. Shelton, C.H., R.D. von Bernuth, and S.P. Rajbhandari. 1985. A continuous-application rainfall simulator. Trans. ASAE 28: 1115–1119. Soil Science Society of America (SSSA). 2006. Soil science terms glossary [Online]. Available at http://www.soils.org/sssaglo ss/index.php. (verifie d 5/16/06). SSSA, Madison, WI. Soil Survey Staff. 2003. Soil Survey of Okeechobee County, FL. U.S. Gov. Print. Office. Washington D.C. South Florida Water Management Dist rict (SFWMD). 1997. Surface water improvements and management plan – Update for Lake Okeechobee. Planning document. West Palm Beach, FL. Swanson, N.P. 1965. Rotating boom rainfall simula tor. Trans. the Amer. Soc. of Agric. Eng. 8: 71–72. Taylor, G.J., A.H. Adriano, and A.H. J ohnson (ed.). 1989. Acidic Precipitation: Biological and Ecological Effects. Springer Verlag, New York, NY. pp. 327–361. United States Department of Agriculture (U SDA). 1990. State soil geographic database for Florida. U.S. Dept. of Agric., So il Conservation Service, Gainesville, FL. U.S. Environmental Protection Agency (US EPA). 1988. Test methods for evaluating solid waste. Volume IC: Laboratory Manual Physical, Chemical Methods. 3rd ed., PB 8-239223, SW-846. Off. Solid Wast e Emerg. Resp., Washington, D.C. U.S. Environmental Protection Agency (US EPA). 1993a. Methods for determination of inorganic substances in environm ental samples. 356.1. Revision 2.0. U.S. Environmental Protection Agency (US EPA). 1993b. Test methods for evaluating solid waste, physical/chemical methods. Method 3050A. 3rd ed. U.S. Environmental Protection Agency (US EPA). 1993c. Test methods for evaluating solid waste, physical/chemi cal methods. Method 350.1. 3rd ed. Van Wambeke, A. 1976. Formation, distributi on, and consequences of acid soils in agricultural development. In: Plant adap tation to mineral stress in problem soils (Ed. M.J. Wright and S.A. Ferrari). Cornell Univ. Agric. Exp. Sta. pp. 15–24.

PAGE 71

60 Withers, P.J.A., S.D. Clay, and V.G. Br eeze. 2001. Phosphorus transfer in runoff following application of fertilizer manure and sewage sludge. J. Environ. Qual. 30: 180–188. Yu, B., C.A.A. Ciesiolka, and P. Langford. 2003. Calibration of oscillating nozzle-type rainfall simulator. Earth Surf Process. Landforms 28: 1483–1490. Zhang, G.C., and G.J. Taylor. 1988. Effect of aluminum on growth and distribution of aluminum in tolerant and sensitive cultiv ars of Triticum-Aestivum L. Comm. Soil Sci. Plant Anal. 19: 1195–1205.

PAGE 72

61 BIOGRAPHICAL SKETCH Thomas J. Rew (“TJ”) was born on August 15, 1981. He is the first child of Tom and Diana Rew. TJ grew up in a military family and attended eight schools before earning his undergraduate degree from the Un iversity of Florida in Environmental Management in Agriculture with a Land a nd Water Specialization (August 2003). While a high school and undergraduate student, TJ spent his summers and holidays working on the family cattle ranch, Hayman’s 711 Ranch. It was here he developed a love of the outdoors. This experience led TJ to develop an interested in the interdependency of the environment, agriculture, and the world’ s growing population. In August of 2003, TJ began his master’s work in soil and wate r science, focusing on phosphorus loss from manure impacted soils, at th e University of Florida.