The effects of harvest and site preparation on the nutrient budget of an intensively managed southern pine forest ecosystem


Material Information

The effects of harvest and site preparation on the nutrient budget of an intensively managed southern pine forest ecosystem
Physical Description:
xii, 185 leaves : ill. ; 28 cm.
Burger, James Anthony, 1945-
Publication Date:


Subjects / Keywords:
Forest soils -- Florida   ( lcsh )
Forest management -- Florida   ( lcsh )
Pine -- Florida   ( lcsh )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis--University of Florida.
Includes bibliographical references (leaves 172-184).
Statement of Responsibility:
by James Anthony Burger.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 06419175
System ID:

Full Text








To my Mother
and the memory
of my Father


I would like to thank all the people who have assisted me during

the course of this research. In particular, I wish to acknowledge the

valuable advice and guidance of Dr. W. L. Pritchett, my committee

chairman, and the assistance from my committee members, Dr. W. H.

Smith, Dr. F. G. Martin, and Dr. D. F. Rothwell. I also wish to

thank C. Burger, M. McLeod, D. Boomsma, and W. Garbett for their

technical services and encouragement.

Many other people have helped at various stages of this work both

in the field and in the laboratory. My thanks go to D. Duncan,

L. Morris, S. Patterson, W. Waller, and L. Shahrabani.

I would also like to thank Reah Duncan for typing this paper and

give a special thanks to my family.

I wish to acknowledge and express my gratitude to the Cooperative

Research in Forest Fertilization program for financial support.



ACKNOWLEDGEMENTS.................................................. iii

LIST OF TABLES..................................................... vi

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

ABSTRACT.......................... ................................ xi

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

LITERATURE REVIEW.................................................. 4

The Ecosystem Approach for Evaluating Impacts on Forests...... 4
Intensive Forest Management................................... 6
Site Preparation.............................................. 9
Mineral Cycling in Forest Ecosystems.......................... 16
Impacts of Intensive Management on Mineral Cycling............ 21
Forest Floor Dynamics... ................................ .. 26
Nitrogen Mineralization.......................... ... ....... .. 31

MATERIALS AND METHODS...... ........................................ 36

Study Area Description and Characterization................... 36
Control Stand Installations................................... 45
Harvesting, Site Preparation, and Planting.................... 50
Harvested Area Installations................................. 52
Laboratory Mineralization Study............................... 56

RESULTS............................................................ 61

Site Characterization......................................... 61
Effects of Harvesting and Site Preparation.................... 71
Stand Establishment............................................ I

DISCUSSION........................................................ 118

Site Characterization........... ..... ...... ....... ... ..... 118
Effects of Harvesting and Site Preparation.................... 124
Stand Establishment... ............. .............. ......... 145

SUMMARY AND CONCLUSIONS............................................. 147


APPENDICES ........................................................ 154


B SOIL PROFILE DESCRIPTIONS......... .... .... .... ... ....... 168

LITERATURE CITED......................................... .... 172

BIOGRAPHICAL SKETCH... .......................................... 185


1. Nutrient content and fluxes in ecosystem components of
three forest types. 19

2. Physical and chemical properties of the two soil types
present on the study area. 44

3. Layout for stemflow collection: Number and diameter of
trees within ten 30 m2 circular plots on two soil types. 49

4. Biomass and chemical element content of the vegetation,
forest floor, and soil components of the ecosystem. 63

5. Average concentration of selected elements in precipitation,
throughfall, and stemflow. 65

6. Nutrient content of the various measured input components
for the year beginning May 18, 1977. 66

7. Concentration of selected elements in litterfall over a one
year period. 69

8. Biomass and content of selected elements in litterfall
collected over a one year period. 70

9. Comparisons of soil properties between treatments and among
years since time of site preparation. 72

10. Effect of treatment on the total organic matter and nitrogen
content of the litter and surface soil (0 20 cm). 74

11. Effects of soil type and site preparation on soil solution
pH and concentrations of selected elements. 78

12. Effect of burning on soil solution pH and levels of selected
elements. 85

13. Soil solution pH and concentrations of selected elements at
various soil depths. 90

14. Effect of time since clear-cut harvest (growing season) on
soil solution pH and concentration of selected elements. 91

15. Effects of site preparation and time since clear-cut harvest
(growing season) on selected nutrient elements. 93

16. Effects of liming and harvest and site preparation on
cumulative amounts of N mineralized (NH4-N + NO3-N) over
an 18-week period. 99

17. Effects of harvest and site preparation on mineralization
potential and rate and adjusted and unadjusted mineraliza-
tion potential half-life. 105

18. Effects of site preparation intensity on survival, height,
diameter, volume and foliage concentration of N, P, and K
of slash pine seedlings after two growing seasons. 113

19. Changes in growth and foliar N, P, and K concentrations
with time. 113

20. Effect of site preparation method on total biomass of
competing vegetation. 114

21. Element concentration of competing vegetation. 117

22. Effects of site preparation method on biomass and nutrient
content of competing vegetation. 117

23. Nutrient content and fluxes of ecosystem components of a
slash pine forest. 120

24. Estimates of output fluxes of selected nutrient elements
as affected by harvesting and site preparation. 137

25. Estimates of N, P, and K storage and flux in ecosystem
components as affected by harvest and site preparation. 139

26. Stand biomass and nutrient content and their distribution
among tree components. 156

27. Nutrient concentrations of tree components. 156

28. Biomass and nutrient content and concentration of the
understory vegetation. 157

29. Concentrations of selected elements of the 01 and 02
horizons on two soil types. 158

30. Stocking rates for a mixed slash-longleaf pine stand in
northern Florida. 159

31. Analysis of variance design used to analyze soil properties
as influenced by soil type and horizon. 159

32. Biomass and chemical element content of the 01 and 02
horizons on two soil types. 160

33. Summary of F probability level of analyses of variance
for all measured soil properties as influenced by soil
type and horizon.

34. Analysis of variance design used to analyze soil bulk
density, organic matter content, C/N ratio, total N, and
extractable P, K, Ca, Mg, Fe, Al, and Na as influenced by
soil, site treatment and time.

35. Effect of treatment on soil temperature and moisture at
four soil profile levels.

36. Analysis of variance design
pH, NH4-N, N03-N, P, K, Ca,
fluenced by soil type, site
depth, and growing season.

37. Analysis of variance design
pH, NH4-N, N03-N, P, K, Ca,
fluenced by soil type, site
growing season.

used to analyze soil solution
Mg, Fe, Al, and Na as in-
preparation treatment, soil

used to analyze soil solution
Mg, Fe, Al, and Na as in-
preparation treatment, and

38. Analyses of variance: Test of the effects of site treatment
on transformed NO and k values.

39. Analysis of variance design used to evaluate the effects of
soil type, site treatment, burn, and season on survival
growth, and foliage nutrient concentrations of slash pine.



1. Vegetation map of the study area. 39

2. Soil horizon depth and organic matter content of the soils
found on the study area. 42

3. Average monthly precipitation for the Gainesville, Florida area. 46

4. Throughfall, stemflow, and litterfall collection devices
used in the study. 48

5. Soil map and experimental layout of the study area. 51

6. Illustration of a soil core used in the N mineralization
study. 58

7. Soil core leaching apparatus used in the N mineralization
study. 60

8. Periodic litterfall in a mature slash-longleaf pine stand. 68

9. Effects of harvesting and site preparation on water table
depth. 76

10. Effects of harvesting and site preparation on soil solution pH. 79

11. Effects of harvesting and site preparation on soil solution
NH4-N concentration. 80

12. Effects of harvesting and site preparation on soil solution
NO -N concentration. 81

13. Effects of harvesting and site preparation on soil solution
P concentration. 82

14. Effects of burning, on plots that were clear-cut and chopped,
on soil solution pH. 86

15. Effects of burning, on plots that were clear-cut, bladed,
disced,and bedded, on soil solution pH. 87

16. Effects of burning, on plots that were clear-
cut and chopped, on soil solution NO3-N concentration. 88

17. Effects of burning, on plots that were clear-cut, bladed,
disced,and bedded, on soil solution NO -N concentration. 89

18. Titrations of two soil samples representing the range of
buffering capacity. 95

19. Millequivilents of Ca(OH)2 needed to adjust the soil pH
back to field levels after leaching with CaCl2. 96

20. Change in soil [H+] as a result of leaching limed and
unlimed soil with 0.01 M CaCl2. 97

21. Effects of harvesting and site preparation on cumulative
N mineralized with time. 100

22. Effects of harvesting and site preparation on the reciprocal
of cumulative N mineralized with the reciprocal of time. 101

23. Effects of harvest and site preparation on the amount of
mineralizable N remaining with time. 103

24. Effects of harvesting and site preparation on the amount
of mineralizable N remaining with time. 106

25. Effects of harvesting and site preparation on N minerali-
zation over the 18-week incubation period. 109

26. Effects of harvesting and site preparation on N minerali-
zation over a five year period. 110

27. The effect of soil pH on nitrification. 112

28. Growth of competing vegetation on the chopped area at time
of planting and after the first growing season. 115

29. Growth of competing vegetation on the intensively prepared
area at time of planting and after the first growing season. 116

30. Nutrient and water storage components and fluxes studied in
a slash-longleaf pine ecosystem. 125

31. Nutrient and water storage components and fluxes of a
clear-cut slash-longleaf pine ecosystem. 127

32. Nutrient and water storage components and fluxes of a
clear-cut and site prepared slash-longleaf pine ecosystem. 127

33. Effects of treatment intensity on the amount of mineral N
provided to a stand during the course of a rotation 144

Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy


James Anthony Burger

August 1979

Chairman: William L. Pritchett
Major Department: Soil Science

The effects of harvest and site preparation on the nutrient cycle

of a slash-longleaf pine (Pinus elliottii Engelm.-Pinus palustris Mill.)

ecosystem were determined in order to evaluate the impacts of harvesting

and two intensities of site preparation on the systems's productivity. A

nutrient base-line was established by characterizing nutrient reserves

stored and cycled in the biomass, soil, and hydrologic components of the

steady-state system. Removals of nutrients in biomass, changes in soil

properties and soil solution composition, N mineralization potentials,

water table level, and input-output fluxes were evaluated for two years

after harvest. Differences in growth of the succeeding slash pine planta-

tion due to intensities of site preparation were evaluated.

Intensive site preparation, which included burning, blading, discing,

and bedding, resulted in a significantly larger amount of nutrients re-

moved from the site than site preparation consisting only of burning and

chopping. A conventional bolewood harvest removed 54 kg N/ha or 13% of

the above-ground N. Blading removed an additional 55 kg of N in the

slash, 44 kg in the understory, and 263 kg in the forest floor. Discing

and bedding caused an increase in the level of all measured nutrients in

the soil solution; all but NH4-N, P, and K returned to steady-state levels

by the second year. During Spring, Summer, and Fall of the first year an

increase in pH was followed by an increase in nitrification due to the

effects of slash burning. Depth to water table level decreased with in-

creasing preparation intensity.

Soluble carbon, extracted from soil surface and Al horizon organic

reserves two years after site preparation, showed that proportionately

more stable reserves remained with increased harvesting and site prepara-

tion intensity. Nitrogen mineralization potential, which accounted for

both quality and quantity of the organic reserves, was used to gauge im-

pacts on site productivity. Nitrogen mineralization potential was not

affected by harvesting and chopping; however, it was significantly reduced

by more intensive site preparation. Although total N reserves and

mineralization potential were lower on the intensively prepared site, when

adjusted for field temperature and moisture conditions, the mineralization

rate was nearly as high as on the chopped-only site and control area.

Without further organic inputs into the soil, projections showed that

most mineralizable N would be depleted by the end of two years.

Intensive site preparation resulted in enhanced tree growth through

better moisture conditions, enhanced levels of nutrients, and a reduced

level of competition. Tree volume on this site was nearly three times

greater than that on the chopped-only site.

In spite of better tree growth during the first two years, an analy-

sis of the nutrient budget, as it was affected by site preparation intensi-

ty, showed that nutrient deficiencies could become a problem later in the

stand rotation.


The Southeast has about 81 million hectares of forest land. These

lands are expected to supply a large portion of the nation's future

timber requirements (McClurkin and Duffy, 1975). However, harvesting

followed by inadequate regeneration practices and the diversion of

forest lands to other uses have resulted in an average annual loss of

roughly 200,000 ha (Knight, 1977). In order to keep pace with

these losses intensive forest management is being practiced, particular-

ly by industrial landowners, to improve the productivity of the

diminishing forest-land base.

Intensive forest management practices common to the flatwoods pine

forests of the lower coastal plain involve the clear-cut removal of the

tree overstory, reduction of the shrub and herbaceous understory, and

mechanical disturbance of the soil in an attempt to optimize space,

moisture, temperature, aeration, and light. Site productivity is

enhanced by rapid establishment and early growth and by good survival,

all of which help ensure good stocking.

Harvesting and site preparation are, however, cultural practices

with inherent dangers, and forest lands are susceptible to abuse if

these practices are not carefully prescribed. While increasing site

productivity through better establishment techniques, there is increased

concern that a net loss in productivity may be incurred in the long-term

due to the inability of the site to provide the nutritional environment

that fast-growing, select trees, grown in shorter rotations, will

require. Reported declines in productivity in other parts of the world

have prompted concern that similar declines might occur here in the


The southern pine ecosystem, located on acid sands of the lower

coastal plain, prospers as the result of conservative nutrient cycles.

These sandy soils are low in nutrient reserves, most of which are con-

tained in the organic fraction. It appears that much of the nutrient

supply of established pine stands is obtained directly from forest

floor materials. Removal of this nutrient reserve, and the subsequent

disturbance of the remaining substrate, constitutes a dramatic change

in the site's nutrient status. Sites located on deep sands having a

relatively small proportion of available nutrients in the soil compared

with the total biomass could eventually be depleted of nutrients by

using these methods.

Increased biological activity after soil disturbances increases

the likelihood of losses by leaching and denitrification from wet,

sandy soils. Fertilization with inorganic materials can compensate for

these losses, but they are subject to leaching and may contaminate

ground and surface water supplies. Significant growth responses have

been obtained through the use of fertilizers; therefore, their use

should be encouraged where needed. However, careful management of

existing organic nutrient supplies is advisable in all instances.

Forest management practices should be geared around nutrient conserva-

tion and environmental quality.

A careful analysis and evaluation of the site's nutrient status

prior to harvest, and for the first several years after harvest and

subsequent site preparation, should provide better projections of long-

term effects of cultural practices on site productivity. The analysis

of such a problem, if studied within the framework of an ecosystem con-

cept, provides a sound foundation for investigations designed to eluci-

date the functional processes involved and show their relation to forest

productivity. This dissertation reports results of a field and

laboratory study initiated to investigate these processes. The

objectives of this study were:

1. Determine steady-state nutrient storage and flows of a

45-year-old slash-longleaf pine (Pinus elliottii Engelm.-Pinus palustris

Mill.) ecosystem.

2. Evaluate the effects of intensive management practices on the

nutrient cycle.

3. Evaluate the effects of intensive management practices on the

growth of the succeeding stand.


The Ecosystem Approach for Evaluating Impacts on Forests

Analysis of ecosystems is based on the premise that it is necessary

to assess the responses of the system to both deleterious stresses and

improved management strategies and to evaluate both direct and indirect

consequences of human intervention (Reichle, 1975). This approach has

been especially emphasized by the productivity subgroups of the

International Biological Program (IBP) (Dale, 1970). These ecosystem

research programs were begun about ten years ago amid doubts that

analysis of ecosystems was technically feasable. According to Reichle

(1975) there was doubt in the broader biological community that the

concept of an ecosystem itself could be a viable approach to resolving

environmental problems. These fears have since been allayed with the

obvious success of the IBP.

Although there were several earlier champions of this approach

(Lindeman, 1942; Sukachev, 1960; Rowe et al., 1960), Ovington (1962)

displayed considerable insight in outlining the requirements for a com-

prehensive study of woodland ecosystems. He recognized that bringing

together fundamental relationships among forest components would pro-

vide an understanding of the dynamic nature of woodlands, not only

throughout the year, but on a long term basis spanning generations of

trees. He suggested that the ecosystem concept would make its greatest

contribution by elucidating the functional processes of woodlands and

the effects of these processes on forest productivity and the possi-

bilities for long-term improvement.

Besides the IBP subgroups, there are many examples of studies in

which an ecosystem approach has been used to meet specific objectives.

Most studies include the characterization and/or cycling of one or more

of the following: energy flow, water, minerals, plants, and animals.

In the eastern U.S., the Hubbard Brook ecosystem study in New Hampshire

is an example which includes all of the above components (Likens et al.,

1977; Gosz et al., 1978). The Coweeta Hydrologic Laboratory in North

Carolina and the Walker Branch Watershed at Oak Ridge, Tennessee, among

others, were designed to study vegetation dynamics and the physical,

chemical, and biological processes regulating immobilization, cycling,

and release of mineral elements (Monk et al., 1977; Henderson and

Harris, 1975). The Coweeta study has recently had its objectives ex-

tended to include manipulated systems that will elucidate effects of

current levels of timber utilization on sustainable productivity (Waide

and Swank, 1977).

Other studies have been initiated with the sole purpose of assessing

effects of intensive management practices on the ecosystem (Smith and

Swindel, 1978). There are ecosystem studies that have been undertaken

on systems as diverse as the Artic tundra and subtropical mangrove

(Lugo, 1970). With few exceptions, the ultimate desire is to couple

mechanistic processes into systems models which are capable of simulating

the behavior of a whole system. This will allow extrapolation of

knowledge to other systems or extrapolation in time in the same system

(Reichle, 1975). Rather than directly providing optimal solutions,

studies will involve successive approximations (Dale, 1970), taking

years perhaps, with exact solutions being forever hidden in nature.

Intensive Forest Management

The Resource

Forest resources in the U.S. are being put to many diverse uses.

While demands for sawtimber and pulpwood are always increasing, new

demands for products such as wood chemicals and energy (fuel) are

expected to grow as the country's needs change. Of the two major

timber producing areas in the U.S., the Pacific Northwest and the

Southeast, the latter has greater potential for increasing its present

level of production. This is due to its favorable climate and physio-

graphy which are conducive to the establishment and maintenance of

large-scale domesticated forests (Stone, 1973).

The Southeast has about 81 million hectares of forest, and these

lands are expected to supply a large portion of the nation's future

timber requirements (McClurkin and Duffy, 1975). In 1970, for example,

it supplied 42% of the roundwood for the U.S. (USDA, 1978) and the

percentage is expected to increase. Much like oil, minerals, and

renewable resources such as food, wood will be subject to increased

international trade as populations expand and resources are depleted.

Consequently, recently proposed exports of significant volumes of wood

from the Southeast to countries such as Italy and Sweden will also

affect demand (Zobel, 1977).

Although the Southeast is extensively forested, the current quality

and age of most of its forests may not be sufficient to meet future

demands (Southern Forest Resource Analysis Committee, 1969). Some

of the most pressing problems according to Zobel (1977) include:

(1) the compromising of tree size and quality by cutting at too early

an age because of pressures for an immediate and continuous supply of

timber; (2) in some areas severe damage is being inflicted to the pro-

ductivity of the land by logging at the wrong time with ill-suited

equipment; (3) poorly stocked and scattered stands in dire need of re-

generation are being passed by; and (4) wood is not being completely

utilized with current harvesting techniques. This last problem could

be the result of economic constraints, or, as Miller (1978) suggests,

one of inadequate technology.

Perhaps the greatest potential for increasing the future wood

supply in the South lies with the non-industrial private landowners

(SFRAC, 1969). Zobel (1977) estimated that these landowners have 70%

of the productive timberland in the South and are now producing less

than half their potential. After selling their stands, poor regenera-

tion, or the total lack of it, occurs despite numerous well planned

programs by government and industry. The landowners have not been con-

vinced that their forest land is a valuable asset worthy of investment

(Zobel, 1977; Miller, 1978). Government policy is yet another considera-

tion when attempting to project wood supply. The future effect is

uncertain, but considering pending land use and environmental quality

legislation, loss of productive forest lands can be expected (McClurkin

and Duffy, 1975; Merrifield, 1978).

Increasing Timber Supplies

Although adequate incentives to private nonindustrial landowners

and better wood utilization at harvest and at the mill would do more

for increasing wood supplies, the option generally being applied,

primarily by industrial landowners, is improvement of land productivity

(McClurkin and Duffy, 1975; Zobel, 1977). Even-aged management of

fast growing pine species in plantations is already common practice.

The advantages and consequences of forest monoculture were reviewed by

Chaiken (1961). In addition to even-aged management, productivity is

being increased with intensive site preparation, land drainage, use of

genetically improved seedlings, fertilization, chemical release,and

full stocking (Bengtson, 1978).

Balmer and Williston (1975) maintain that a forest owner who wants

to produce the largest yields and maximize returns on investments must

practice intensive stand management. An extensive study of economic

management opportunities to increase timber supplies in the Southeast,

undertaken by the U.S. Forest Service and the Forest Productivity

Committee of the Forest Industries Council, confirms this assessment

(Dutrow, 1978). The objectives of that study were to determine (1) what

increases in national timber supplies might be achieved by applying more

intensive cultural treatments on commercial forest lands; and (2) if

these cultural treatments promise an acceptable return on investments.

Investments ranged from only a few dollars per acre in stand improvement

to $115.00 per acre for widespread clearing and regenerating. The

results showed that slight expenditures to improve existing stands were

more rewarding than costly conversions to bare ground, but plantations

on converted land promised highest yield increments and good returns.

The fact that conversion options from low quality hardwoods to pine

included two-thirds of the estimated area needing some form of treatment

ensures that intensive treatments, including clearing, site preparation,

and planting, will be an increasingly used option for pine production.

Merrifield (1978) emphasized that given enough productive sites for

forest production, it is our stewardship of the soil that will largely

determine our ability to produce enough fiber.

Site Preparation


Prior to World War II nearly all site preparation was done with

hand tools (Brown, 1971). Chavasse (1974) reported that there were only

eighteen references in Forestry Abstracts to papers dealing with site

preparation machinery during the period 1960 to 1964. During 1965 to

1970, eighty-four references were recorded. Site preparation with

heavy machinery is a relatively recent phenomenon, but today it is

widely used in all mechanized countries. Large management areas and a

limited supply and high cost of labor are the primary reasons for this

increased mechanization.

An extensive review of site preparation for stand development was

made by Post (1974). He reviewed techniques and biological considera-

tions of the treatments by world regions including Scandinavia, Central

and Western Europe, Soviet Russia, New Zealand, Australia, Canada,

Africa, and the United States. A more detailed review was made by

Chavasse (1974) who discussed objectives, current trends, methods, and

environmental consequences of site preparation. Brown (1971) reviewed

site preparation in Australia where extensive conversion of eucalypt

woodlands to Pinus radiata are being made. A summary of site prepara-

tion in the Pacific Northwest was compiled by Stewart (1978).

These authors considered site preparation priorities, methods, factors

influencing choice of method, and finally, based on their experience,

suggested site preparation methods for specific sites.

Site preparation has long been practiced in the Southeast,

especially by forest industries that were instrumental in the development

of many of the techniques. Excellent reviews of the subject for the

Southeast region have been made (Wilhite, 1961; Haines et al., 1975;

Terry and Hughes, 1975), and other descriptions of techniques and methods

are available (Wilhite, 1961; Shores, 1968; Pritchett, 1969; Balmer

et al., 1976; Balmer and Little, 1978).

Intensive land clearing and subsequent cultivation is often a

requirement of intensive production forestry. The general aim of soil

preparation is to ensure the establishment of fully stocked, actively

growing, uniform stands at minimal cost and to facilitate the conduct

of subsequent operations (Brown, 1971). More specific aims include

(I) improve survival by removing competing vegetation and overhead

shade; (2) improve early soil moisture conditions by eliminating com-

peting vegetation and excess water; (3) facilitate machine planting by

eliminating cull trees or logging debris; (4) improve wildlife food and

cover, and reduce fire hazards; (5) improve early tree growth; (6) pre-

pare a mineral seedbed; (7) reduce compaction and improve internal

drainage; (8) control disease and insects; and (9) improve availability

of nutrients, especially N, by breaking down and incorporating organic

matter into the soil (Haines et al., 1975; Balmer et al., 1976; Balmer

and Little, 1978).


Fire, machinery, and chemicals are the usual methods used to

achieve the objectives listed above. Fire, perhaps the most important

and least expensive tool, reduces logging debris and the aerial portions

of competing vegetation, thereby facilitating planting, reducing fire

hazards, and partially removing completion to pine stands (Wilhite and

Harrington, 1965). Because of air pollution problems associated with

burning, local and/or state laws may prevent burning during part or all

of the year. The effect of particulates from this source on general

air quality is not well defined (Ward and Elliot, 1976). In spite of

its low cost and many benefits, burning alone often does not adequately

prepare the site. Large pieces of debris are not consumed and advanced

hardwood competition is often not deadened (Wilhite and Harrington,

1965). In addition to the large debris, high stumps make machine

planting difficult or impossible.

Because of vast acreages of potential pine land now in low quality

hardwoods, hardwood scrub, and poorly stocked natural stands of southern

pine, extensive land clearing with KG blades and root rakes will con-

tinue to be common. Conversion of unmerchantable eucalypt stands in

Australia (Brown, 1971), clearing for Douglas-fir plantations in the

Pacific Northwest (Stewart, 1978) and the removal of low quality

hardwoods in preparation for pine in the Southeast (Balmer and Little,

1978), are examples of past and current uses of this method. Pritchett

(1979) described the operation as involving large dozers equipped with

KG blades or root rakes heaping trees and logging debris into windows

spaced 60 to 100 m apart. For slopes greater than 15, windows are

aligned with the contours to reduce runoff and erosion. After burning

the windows, 5 to 20% of the total space may be unplantable by machine.

This operation is probably the most detrimental of all methods because

it removes the forest floor, humus, and as much as 5 cm of the Al soil

horizon (Haines et al., 1975).

Harrowing with heavy tandem disc harrows is most often undertaken

after first clearing the site. It is especially effective against

vegetation which develops a dense root mat just below the soil surface,

i.e. saw palmetto (Serenoa repens (Bartram) Small), wiregrass (Aristida

stricta Michx.), gallberry (Ilex glabra L. Gray), and runner oak

(Quercus pumila Walter) (Haines et al., 1975). Where KG blading is not

necessary, harrowing can be used to break up a heavy forest floor to

expose mineral soil. Effects are enhanced if harrowing is done one or

two years after clear-cutting to allow time for rotting of logging

slash and stumps (Wilhite, 1961).

Bedding is a standard soil preparation technique used primarily in

wet areas of the coastal plain. Successes in the Southeast on pre-

dominantly sandy soils, however, led to studies further west on finer

textured, siltier soils common in the flatwoods of southwest Louisiana

and southeast Texas (Derr and Mann, 1970; McKee and Shoulders, 1974).

Bedding concentrates surface soil, litter, and logging debris into a

ridge 15 to 20 cm high and 1 to 2 m wide. It has much the same effect

as harrowing in that conditions for accelerated decomposition of

nutrients are enhanced. Beneficial effects have been attributed to

improved drainage where long periods of soil saturation reduce aeration

in the rooting zone (Bethume, 1963; Balmer et al., 1976; Derr and Mann,

1977). Other benefits include improved nutrition and reduced competition

(Haines and Pritchett, 1964; Haines and Pritchett, 1965; Shores, 1968;

Schultz and Wilhite, 1974; Terry and Hughes, 1975).

Results of Site Preparation Treatments

The increase in survival, early growth, and overall success in

pine establishment due to site preparation is well documented. A

number of investigators working on wet sites in the Southeast reported

gains in early growth of loblolly and slash pine as a result of harrow-

ing and/or bedding on cleared sites (Langdon, 1962; Bethume, 1963;

Worst, 1964; Schultz, 1973; Mann and McGilvray, 1974; Pritchett and

Smith, 1974; Terry and Hughes, 1975; Derr and Mann, 1977).

Terry and Hughes reported that a 13-year-old study of bedding

versus flat planting had slash pine trees with heights of 9.7 and 7.9 m,

respectively. Total volumes (inside bark) of trees on the two treat-

ments were 68.5 versus 41.1 m3/ha in favor of the bedded treatment, a

difference they felt would be long lasting. Worst (1964) reported an

average of 36% greater height of four-year-old slash pine on burned and

bedded plots than on nonprepared plots. Greater height differences

diminished with time and only on the poorly drained soils were volume

differences significantly greater than the control.

Derr and Mann (1977) report that six studies of bedding on poorly

drained sites in southwest Louisiana have not provided conclusive

evidence that this intensive treatment is worth the extra cost. There

was a significant boost in height growth in four of the six studies but

the maximum response was 0.8 m of added height at age five. At age 10

the increase was only about 0.6 m, but they maintained that it was too

soon to determine if early gains will entirely diminish with age.

Bedding increased the incidence of fusiform rust by 5 to 50% in the Derr

and Mann study, but the increased infection was not thought to have

much economic effect. Preparation treatments applied prior to bedding

were not reported.

There have been additional studies that indicated treatment effects

may not be long lasting. Lennartz and McMinn (1973) found that bedding

significantly improved growth in four of six plantings after five years,

but by the tenth year only on one test had bedding maintained a signifi-

cant increase in growth rate. Haines et al. (1975) reported on a north

Florida study that showed a clear reduction in growth of slash pine on

beds after an initial gain on trees planted on flats. Height differ-

ences increased in favor of the trees on the beds until age 8.5 and

then decreased until about 14.5 years when the difference was only one-

third as great as at 8.5 years. The diameter differences were greater

at 6.5 years but declined afterwards until trees on the flats were

slightly larger than those on the beds by age 14.5 years. Lennartz

and McMinn (1973) attributed early growth and subsequent decline to

improved soil moisture conditions and better aeration, but as the root

systems of trees in all treatments began to fully utilize the site,

they felt that other factors influencing growth became more important.

After about 10 years, nutritional deficiencies could become growth

controlling factors accounting for diminishing gains on bedded or

harrowed sites compared with unprepared areas. Burned sites, on which

the organic fraction is otherwise left undisturbed, may be at the best

nutritional advantage at this point in the rotation because their

nutrient reserves have been largely left intact. Harrowed and/or bedded

sites, on the other hand, have organic fractions that have been mixed

and aerated resulting in the premature decomposition and mineralization

of their most reactive organic matter. Sites that have been cleared by

KG blades or root rakes and have had their organic fraction windrowed

would be in the most nutritionally degraded state. Glass (1976) re-

ported a study on a site located in Durham County, North Carolina,

illustrating the latter condition. Part of an area was clearcut and

burned while a second part was prepared by removing the hardwood root

mat and piling it, along with the forest floor and an estimated 5 cm

of surface soil, into windows. Both areas were planted with loblolly

pine. Nineteen years after establishment, the broadcast burned area

contained 346 m3/ha of wood compared with 187 m3/ha for the root raked

area a 46% reduction in productivity. On the root raked area, the

trees immediately adjacent to the windows were significantly larger in

total and merchantable heights, diameter, and volume than the codominant

trees in the remainder of the plantation. These results are in agree-

ment with those reported by Cromer (1967) that showed ash-bed

or burned window sites produced 50% more merchantable timber after 12

years than adjacent cleared sites. They attributed the growth stimula-

tion to several factors including concentration of all nutrient capital

contained in the original vegetation on one-tenth the area, transfer of

topsoil from the cleared areas to the windows, and the addition of

nutrients from the wood ash. These studies illustrate the value of

maintaining an organic nutrient reserve on the site to tree nutrition

and long-term productivity.

Mineral Cycling in Forest Ecosystems


The productivity of a forest ecosystem depends, to a large extent,

on the system's ability to conserve and recycle its nutrient resources

(Mitchell et al., 1975). Through a dynamic and rather complex system

of geological, chemical, and biological cycling, the soil organic matter

and nutrient supplies are replenished and maintained, thereby ensuring

continued productivity of forest land (Pritchett, 1979). World litera-

ture on mineral cycling and mineral inventories in forests is extensive.

Before the advent of the International Biological Program (IBP),

Europeans were the primary contributors. Rodin and Bazilevich (1967)

reviewed biological mineral cycling in world ecosystems. Less extensive

but more intensive was a review by Ovington (1962) on cycling in hard-

wood forests in Great Britain and a review by Duvigneaud and Denaeyer-

DeSmet (1970) of research in deciduous forests of Belgium. In North

America, Curlin (1968) reviewed the behavior and movement of macro-

nutrients from trees through the biotic phase of the nutrient cycle.

More recent summaries contain studies in which the effects of man's

activities are included in the realm of cycling. In a review by Krause

et al. (1979) mineral cycles in boreal forest ecosystems were charac-

terized, and man's effects on nutrient cycling and productivity in these

systems were evaluated. Much the same approach was used by Ralston

(1979) when reviewing mineral cycling in temperate forest ecosystems

and by Hollis et al. (1979) in their review of mineral cycling as it is

affected by silvicultural practices.


The forest nutrient cycle is basically composed of three segments

which include an input, an intracycle or system within which nutrient

movement takes place, and an output (Jorgensen et al., 1975). The

cycle is also described as two major subcycles: an external geochemical

cycle and an internal biological cycle (Remezov, 1959; Pritchett, 1979).

The geochemical system corresponds to the input and output segments

while the biological cycle is the same as the intracycle segment. Each

approach has its advantages for studying and evaluating cycling, but

the former is usually used in the context of ecosystem analysis, with

system boundaries around the intracycle and inputs and outputs crossing

the boundaries.

Nutrients are found in three compartments within the system: (1) in

organic materials, (2) in the pool of available nutrients in the soil,

and (3) in soil and rock minerals (Bormann and Likens, 1967). Inputs

to the system are from dissolved gaseous or solid materials in pre-

cipitation, particulate matter that settles on the site, by fixation of

molecular nitrogen by soil microorganisms, and by geologic weathering

of primary minerals (Jorgensen et al., 1975).

Nutrients can escape the system by dentrification, leaching, and

surface runoff. Important inputs and outputs in managed systems are

fertilization and loss by burning and harvesting. Geochemical nutrient

cycling has been comprehensively reviewed by Pritchett (1979). In

general, inputs will balance any losses from the system, but whether a

net loss or gain is observed will depend on the type of system and its

current progress toward successional equilibrium and the nutrient

element in question. Likens et al. (1977) made extensive studies of

input-output budgets for several temperate forest ecosystems. The

results, subsequently summarized by Ralston (1979), showed that for 15

observations the systems gained an average of 5.9 kg N/ha, 0.11 kg P/ha,

and lost 1.9 kg K/ha per year.

Nutrient transfers among ecosystem compartments (biological cycling)

involve the uptake of mineral elements, their storage in perennial parts,

and the return of a substantial fraction of these elements to the

forest floor by way of throughfall, stemflow, and litterfall. The

forest floor supports a variety of heterotrophic organisms which decom-

pose organic matter and convert nutrients to soluble forms suitable for

reabsorption by the vegetation (Ralston, 1979). Biological studies

attempting to quantify the production, annual partitioning, and trans-

fers of minerals among ecosystem components are the subject of recent

reviews by Krause et al. (1979) on boreal forest ecosystems, Ralston

(1979) on temperate forest systems, and Golley (1975) on tropical

forests. Comprehensive mineral cycling studies of coniferous species

have been made by Switzer et al. (1968), Jorgensen et al. (1975), Foster

and Morrison (1976), Webber (1977), and many others.

Nitrogen, P, and K contained in the biomass reserve and moving as

inputs, outputs, or within the intracycle of three forest types are

shown in Table 1. The relative proportions of nutrients in each stand

component and the rate at which they cycle contrast sharply among the

three forest types. The total N in the stems of the loblolly pine stand,

for example, is over twice that of the jack pine stand even

though it is 14 years younger. Uptake of N is 4.5 times greater

in the loblolly stand compared with the jack pine stand but retention

is about the same. This shows the comparatively rapid rate at which

Table 1. Nutrient content and fluxes in ecosystem components of three
forest types.

Forest type Jack pine Douglas-fir Loblolly pine
Location Ontario Washington North Carolina
Age 30 years 37 years 16 years
Source Foster and Cole et al. Jorgensen et al.
Morrison (1976) (1967) (1975)
------------------------- kg/ha ------------------------

foliage and
forest floor
mineral soil/




4216 71.8 508



142 16
115 15
------- NA
307 30
1753 751

469 2317 812




--------------------- kgha 1yr-

7.9 0.1 4.0 --- NA ------
NA NA 4.0a ----- NA ------

5. Id/




NA -------



5.5 0.2 1.5
1.0 NA NA




----- NA ----- 0.7 0.1 1.6
----- NA ------ -------- NA----

/Source: Krause et al. (1979).

ISoil depth: Jack pine 30 cm;
Loblolly pine 70 cm.

Uptake = retention + return.

Douglas-fir NA;

lRetention = (foliage + branches + stem)/stand age.



nutrients cycle in southern pine ecosystems. Greater proportions of

the total nutrients can be retained and kept cycling in the vegetation

which makes the. southern forest types potentially more productive than

northern forest types. Increased productive potential, however, is

traded for mineral-cycling stability. Relative stability of a forest

mineral cycle is a function of the time required for the mineral-

cycling system to recover from a perturbation such as harvest and site

preparation. Recovery time is a function of recycling rate to steady-

state input of material. Mineral-cycling systems with relatively high

ratios of recycling to steady-state input, such as the cycling system

of southern pine, are less stable than systems with lower ratios

(Jordan et al., 1972).

O'Neill et al. (1975) pointed out that certain nutrient elements

are essential for reestablishment of plants; therefore, element con-

servation is required over time spans that exceed those of any

individual population in the ecosystem. Characteristics of a nutrient

element medium should be large size, ability to retain elements, and

slow response time. In forest systems this component is found in the

litter and soil organic matter (Edwards et al., 1974). A comparison of

the total nutrients found in these two components for the forest types

outlined in Table 1 shows that the southern forest system (loblolly

pine) contains considerably less total N than the jack pine or

Douglas-fir forest systems. Any forest practice which greatly decreases

this nutrient reserve could threaten the stability of the mineral-

cycling system as Jordan et al. (1972) suggest, or at least decrease

system productivity.

Impacts of Intensive Management on Mineral Cycling

Silvicultural treatments affect mineral cycling in forest ecosystems

in several fundamental ways: (1) relatively large amounts of the total

nutrient capital of a stand can be removed as a result of shorter

rotations, harvest, burning, and site preparation; (2) nutrients can be

added to a forest by fertilization; and (3) treatments may neither add

nor subtract nutrients from the site, but may substantially alter stand

nutrition by changing the rate of critical processes such as decom-

position and mineralization (Miller et al., 1976).

In an attempt to produce the greatest quantity of utilizable wood

in the shortest time at the lowest cost, most forest stands will be

managed for shorter rotations, especially where the primary product is

fiber rather than solid wood. Switzer and Nelson (1973) examined

changing nutritional relationships as natural forest ecosystems are

converted to short rotation plantations. Theyconcluded that: (1)

shortening the rotation devotes a greater period to regeneration which

increases the exposure of the forest floor to degradation and subsequent

leaching and erosion losses; (2) the frequency of occurrence of these

periods increases; (3) the nutrient demand will be greater because

production per unit area per unit time is greater; (4) nutrient removals

in the crop are greater due to better utilization of the standing crop;

and (5) nutrient drain per unit harvested is greater since nutrient

accumulation is greatestduring the developmental periods of site


Because of the temptation for more complete utilization with short

rotations, an additional burden will be placed on the site's nutrient

reserves at harvest. Harvesting techniques have been developed that

remove stump and central root portions as well as tree crowns (Koch and

Boyd, 1978) which will be processed into products as diverse as fodder

supplements to fuel (Young, 1976; Koch and McKenzie, 1976). Comparisons

of nutrient losses for conventional versus whole-tree harvests have been

made for many forest types. White (1974) reported the effects of whole-

tree harvesting of eight cottonwood stands on the soil nutrient pool.

The data indicated possible site degradation by depletion of soil

reserves of N, P, and K, but not Ca and Mg, on a range of alluvial sites

in Alabama. Leaving foliage on the site still resulted in the removal

of relatively large amounts of P and K. Jorgensen et al. (1975) deter-

mined that a 16-year-old stand of loblolly pine yielded 185 t/ha of

biomass when harvested in its entirety versus 116 t/ha by removing only

the stemwood and bark to an 8 cm top. The latter harvest method removed

one-third as much nutrients, but yielded two-thirds as much biomass.

For a 450-year-old Douglas-fir stand, Miller et al. (1976) estimated

that 242 kg N/ha would be removed by conventional harvest and nearly

double that amount if the crowns were removed.

In contrast to the obvious effects of nutrient removals by har-

vesting, the effects of slash burning are more difficult to determine.

It is generally conceded that adverse effects are minimal because when

properly managed, most of the forest floor remains intact and most of

the nutrients of the burned organic matter remain on the site as ash

(Metz et al., 1961; Wells, 1971; Hatch and Mitchell, 1972). Large

amounts of N and S can be oxidized and lost (DeBell and Ralston, 1970;

Horwood and Jackson, 1975). Pritchett and Wells (1978) estimated that

200 to 300 kg N/ha may volatilize in a hot burn for slash removal that

also burns the forest floor. This can amount to a greater loss than

tree removal. There is evidence, however, that N can be partially re-

placed in a relatively short time by N-fixing organisms that have been

stimulated by post-fire conditions (Metz and Farrier, 1971; Jorgensen and

Wells, 1971; Wells, 1971). Burning can also cause a temporary rise in

soil pH when precipitation leaches the ash into the profile (Moore and

Norris, 1974).

Site preparation which includes land clearing or slash removal

with a KG blade or root rake with subsequent burning of windows can

severelyreduce the site's nutrient reserves. This procedure removes

virtually all logging slash including stumps, understory vegetation,

forest floor materials, and a portion of the Al soil horizon (Schultz,

1975). According to one estimate (Hollis et al., 1979), this treatment

can remove up to 350 kg/ha, or 20%, of the total site N reserves, and

up to 20 kg/ha P in coastal plain ecosystems. Harrowing and bedding

operations do not cause further direct removal, but they change the

physical structure and placement of the remaining organic materials.

This mixing reduces soil bulk density and increases aeration (Haines

and Pritchett, 1964).

The combined operations of harvest, burning, and mechanized site

preparation change the site's microclimate. Schultz (1976) reported

increased temperatures and wind movement on a prepared clear-cut in

northeast Florida. The nutrient cycle is directly tied to the hydrology

of the site. After removing the forest cover, the amount of rain

reaching the forest floor can increase by 15% (Helvey, 1972). A greater

input of rain coupled with lower evapotranspiration increases the level

of soil moisture creating a more responsive hydrologic system (Nutter

and Douglass, 1978). As a result, surface runoff and erosion are in-

creased, the water table often rises in level areas, and the level of

subsurface soil moisture is increased. In combination with increased

temperature and soil aeration, a soil medium is provided that is con-

ducive to accelerated decomposition and mineralization of the remaining

organic matter (Miller et al., 1976).

In order to evaluate the combined effects of intensive management

practices, nutrient balance sheets have been constructed. A general

conclusion was that nutrients removed in conventional harvest were

adequately replaced by natural nutrient additions which should maintain

established growth rates for several generations (Boyle and Ek, 1972;

Weetman and Webber, 1972; Jorgensen et al., 1975). Other studies showed

that when rotations are shortened (Switzer and Nelson, 1973), the entire

stand is harvested (Jorgensen, et al., 1975), or a significant portion

of the site's organic nutrient reserves are removed (Hollis et al.,

1979), a net depletion of nutrients will occur. Computer simulations

of the dynamics of forest ecosystems over several rotations have also

shown decreased productivity as a result of nutrient depletion and

interaction of other site factors. Results of simulations by Waide and

Swank (1977) of oak-hickory and loblolly pine forest systems show that

site yields may be slightly increased, unchanged, or substantially de-

creased after several rotations, depending upon the degree of tree

utilization. A simulation model by Penning de Vries et al. (1975)

depicted the sensitivity of yield to certain management practices that

affect the N cycle.

Initial and subsequent effects of harvesting on mineral cycling

vary greatly with site. According to Miller et al. (1976), reductions

in productivity are minimized on sites which have: (1) a relatively

large capital of organic matter in the mineral soil; (2) deep, well-

drained soils with high infiltration rates and moisture holding capacities

which minimize erosion; (3) favorable climatic conditions that compen-

sate for degradation of soil physical and chemical properties; and

(4) rapid secondary succession by a wide variety of plants which re-

establish a nutrient and organic cycle. Soils of the coastal plain of

the Southeast are predominately classed as Ultisols, Inceptisols, and

Spodosols. They are primarily sandy, siliceous soils which are highly

weathered and have low organic matter contents and cation exchange

capacities. Because these mineral soils are inherently infertile,

forest floor materials are essential for nutrient cycling and for the

adequate nutrition of forest stands of the area. Therefore, maintenance

of the nutrient capital associated with the forest floor is paramount to

the maintenance of soil productivity.

There is increasing concern among silviculturists and forest soil

scientists that productivity of southern pine sites may be declining

because of certain intensive management operations. Pritchett and Wells

(1978) warn of the long term consequences that could result from sub-

stantial organic matter removals. Experience in many other parts of the

world is evidence that fears of productivity decline in the Southeast

are not unfounded. There have been numerous reports of decreased pro-

ductivity with successive rotations, most of which have been summarized

in several comprehensive reviews (Lewis, 1967; Bednall, 1968; Evans,

1976). Others have reported on and reviewed the factors responsible for

possible yield declines (Florence, 1967; Lewis and Boardman, 1969).

Declines were often simply attributed to nutrient losses from the site

due to harvest removal, burning, and leaching, but Florence (1967)

suggested declines were due to more fundamental processes and depicted

the complexity of the interrelationships involved. Processes included

the effect of species and species mixtures on their sites, the impor-

tance of energy flow in the decomposition of litter and return of

nutrients, and the deterioration in the forest microflora relationships.

An understanding of the significance of these processes in natural

forests may help place the plantation problem in better perspective.

Forest Floor Dynamics

Decomposition of forest litter plays a major role in the process

of mineral cycling and tree nutrition. It is a process vital to the

continuing productivity of any forest ecosystem, and within plantation

forestry it must be managed to provide an adequate and timely supply of

nutrients to the forest stand.

The return of organic matter from the vegetation of forest stands

to the soil occurs as litterfall, timberfall, and root decomposition

(Nye, 1961). It is often assumed that litterfall represents all, or

nearly all, the organic matter returning to the forest floor and soil

system, but studies on the other two components reveal that substantial

inputs occur from these sources. McFee and Stone (1966) examined the

amount and composition of decaying wood in the humus layers of yellow

birch (Betula alleghaniensis) and red spruce (Picea rubens) stands in

New York. Separable wood residues weighed as much as 46,850 kg/ha.

Roots can also provide large annual additions to a soil's organic pool

(Harris et al., 1974). Mycorrizalshort roots do not live more than two to

three years, and small non-mychorrhizal roots seldom last over one season.

Quantitative estimates of forest soil organic matter production

and accumulation are difficult to make. Measurements of the forest

floor and subsurface organic matter are subject to large magnitudes of

variation due to soil microrelief, heterogeneity of the forest floor

layer composition, and irregular rooting distribution of higher plants

(Leaf et al., 1971). Pritchett (1979) reported that most conifers and

hardwoods in cool temperate regions return between 2 to 6 t/ha of

litterfall annually, but returns can be as high as 12 t/ha in tropical

rain forests. He provided a summary which reported the weights and

properties of forest floor layers under several forest types. Although

annual returns of organic matter are not very different for a range of

forest types in similar climatic zones, the degree of accumulation in

the forest floor varies greatly (McFee and Stone, 1965; Metz et al., 1970;

Wooldridge, 1970; VanLear and Goebel, 1976). The amount of litter that

accumulates on the surface of the mineral soil depends on the balance

between the rate of litterfall and the speed with which it decays. The

overall rate of decay depends on such factors as degree of lignification

of tissues and prevailing climatic conditions and weather patterns

(Richards, 1976). Decomposition is more rapid in hot climates, but

accumulation is considerably less than in cold climates despite the

fact that annual returns can be as much as twice as high. A convenient

way of comparing rates of organic matter turnover among various forest

types considered to be at a steady state is by dividing the annual

organic matter returns by the amount of accumulation in the forest floor.

The quotient is the decomposition constant. Some decomposition constants

determined in this way are 0.025 for ponderosa pine forests in the

Sierra Nevada Mountains, 0.25 for longleaf and slash pine forests in the

Southeast, and 4.0 for tropical African forests (Olsen, 1963).

Relative contributions of microbes and soil animals to soil

metabolism are unclear. It is generally believed that microflora

(mainly fungi and bacteria) contribute a greater amount to total soil

metabolism than the micro- or macrofauna (Reichle, 1971). Only about

10% of the annual input of energy in litter is thought to be used by

the soil fauna for biosynthesis, growth, and respiration (Richards,

1976). The soil fauna and microflora are symbiotic in the sense that

they produce a synergistic effect on decomposition rates. This is

achieved by particulation of the organic matter by the microfauna which

increases substrate surface area for the microflora to act upon and

facilitate degradation (Spain, 1975). According to Burges (1965), the

decomposition process for pine litter is primarily a fungal and mite

controlled process, although collembola, millipeds, and enchytreids

are also important. In other situations such as deciduous hardwood

and tropical forest types, a much wider range of animals is found. In

the tropics, ants and termites are responsible for the reduction and

burial of a major fraction of the plant debris. In deciduous forests,

larger worms transport plant debris from the surface to the lower layers

of the soil either by direct transport or by ingestion (Burges, 1965).

During decomposition there is a progressive loss of CO2 leading to

a narrowing of the C/N ratio. On the basis of his pine litter studies,

Burges (1965) suggested that there was little loss of N in the L and

Fl layers because plant protein was converted to microbiological protein;

however, at the base of the F2 layer deamination occurs and N is released

as ammonia.

The factors affecting rate of breakdown and the formation of

various types of forest floors are many and their interactions are

complex. The rate of breakdown of a given forest floor is largely a

function of the growth and activities of soil microbes. According to

Bollen (1974), growth and activity are controlled by water, temperature,

aeration, pH, food supply, and biological factors, all of which are so

interrelated that a change in one induces changes in others. Optimum

levels for each of the factors are reported to be 50% of moisture

capacity, 280 C, 50% of soil pore volume, pH 7, C/N ratio of 25, and a

symbiotic condition with limited antibiosis. Factors giving rise to

the formation of a given type of forest floor are more complex.

Handley (1954) evaluated the effects of geology, topography, climate,

soil microflora and fauna, and vegetation and leaf properties. More

recent studies have included enzyme inactivation by tannins and lignin

concentrations as controlling factors (Benoit and Starkey, 1968;

Meentemeyer, 1978).

A most important feature of the forest floor is its role in pro-

viding a revolving fund of nutrients, particularly N, P, and S, for

higher plants (Pritchett, 1979). After organic matter decomposition,

mineral nutrients are transferred to plant roots. In southern pine

ecosystems only negligible amounts of these minerals are released to

the soil. Instead, they are quickly absorbed by roots and transported

to the photosynthetic tissues (Gamble, 1971). Here they are again in-

corporated into organic matter, part of which is retained and part of

which is returned to the forest floor. If this equilibrium is altered,

patterns of nutrient element distribution and turnover rates are also


Cultural treatments, such as the incorporation of various levels

of harvesting residue, additions of fertilizers, and the use of fire and

site preparation, can markedly alter these patterns (Hamilton, 1965;

Florence, 1967; Miller et al., 1976). These changes may be beneficial or

detrimental depending on region, climate, and forest type. In Douglas-

fir and most other northern forest types, natural disposal of forest

residues by any method other than burning usually requires many years.

Therefore, faster decomposition is being stressed (Bollen, 1974; Miller

et al., 1976). Methods that prolong the coincidence of temperature and

moisture ranges favorable to decomposition are favored. Crushing,

chopping, and burying residue are carried out to increase contact with

the soil and provide a moisture retaining mulch (Schimke and Dougherty,

1966). Investigations on accelerating decomposition of Douglas-fir

residue by treating with nitrogenous chemicals have also been performed

without success (Ward, 1975). Miller et al. (1976), however, reported

results showing decreased N residence times of 25% in the forest floor

with application of ammonium nitrate to a forest in the Oregon Coast


In the Southeast the opposite problem of too rapid decomposition

is associated with certain cultural practices. Plant nutrients are

released before an adequate vegetational cover is established, thereby

creating the risk of loss by leaching. It would be advantageous,

therefore, to slow decomposition to some extent, resulting in a more

steady supply of plant nutrients (Pritchett and Wells, 1978).

Decomposition of forest floor residues is particularly dependent

on moisture, temperature, particle size, and sufficient nutrients. The

closer each of these properties approaches optimum, the faster will be

the rate of decomposition and nutrient release (Bollen, 1974). If

cultural practices are geared around the natural processes of organic

matter decomposition, a better nutritional balance during the course of

a forest rotation might be maintained.

Nitrogen Mineralization

Biological mineralization is the conversion of immobilized nutrients

to inorganic form by microbial decomposition. Mineralization of

nutrients in organic matter forms the very basis for long-term forest

nutrition especially in terms of tree N requirements. This process has

long been taken for granted in forestry because tree nutrition was seldom

manipulated until relatively recently. Since discovering that most

forest ecosystems respond to inputs of additional N, attempts have been

made to provide forest stands with optimum amounts of this element in

order to maximize growth. A carefully prepared fertilization prescrip-

tion would require a prediction of net mineralization flow in the N

cycle and a subsequent estimation of the need for supplementary N. The

determination of net mineralization of N and subsequent availability to

higher plants has proven to be extremely difficult compared to successes

in testing for other plant essential macroelements. Consequently, N

availability has been the subject of extensive research, but today N

remains the element used most inefficiently in agronomic and forest

fertilization programs because of its complex transformations in both

biotic and abiotic environments.

A comprehensive review of N mineralization in soils was made by

Harmsen and VanSchreven (1955), and additional summaries have been pro-

vided by Harmsen and Kolenbrander (1965) and Stevens (1965). The

mechanisms of decomposition, immobilization, and mineralization are well

understood, but quantification of the processes has been difficult

because of the many physical, chemical, and biological factors which

affect them.

Nitrogen mineralization and immobilization processes can follow

different paths as outlined by Harmsen and Kolenbrander (1965). The

shortest path is when the organic substrate is partially disintegrated

with the N transformed to ammonium. Part of this ammonium is assimilated

by the microbes or incorporated into humus while a portion is left for

absorption by higher plants. When starting with organic materials with

a high C/N ratio, the carbon is liberated rapidly as CO2 while the N is

retained in organic form. When the energy/N ratio is sufficiently

reduced, inorganic N can accumulate.

C/N ratios have been used extensively to predict levels of N that

could be expected to mineralize. It was generally thought that the

C/N ratio of the decomposing material had to be below 20 to 25 for an

appreciable net mineralization of N (Harmsen and Kolenbrander, 1965).

The dangers of this assumption were pointed out by Bartholomew (1965)

when he considered the relative resistance to decomposition (i.e.,

conifers versus hardwood litter). The general rule that net mineraliza-

tion of organic N depends primarily on the N content of the substrate

holds only for the mineralizable part of the decaying material. This

explains the apparent net mineralization observed in coniferous organic

materials with C/N ratios considerably higher than 25.

The practical value of a method providing an index of the avail-

ability of soil N for accurate prescriptions of fertilizer needs has

long been appreciated by agronomists. Bremner(1965c) reviewed the many

biological and chemical methods that have been proposed. Nutritional

needs in forestry are generally diagnosed on a need--don't need basis,

although in some areas fertilizer recommendations have become more re-

fined (Pritchett and Gooding, 1975). In forestry, indices of N minerali-

zation may have greater application for the evaluation of effects of

intensive management on the efficiency of use of the organic N reserves

over the long term (Tamm, 1975).

Of the many N availability indices reviewed by Bremner (1965c), the

methods involving estimations of the amount of mineral N formed during

incubation have been extensively employed. They are generally considered

to be the most satisfactory methods for assessment of the potential

ability of soils to mineralize N. As Bremner (1965c) points out, this

method has a rational basis because the agents responsible for the N

measured in incubation experiments are the same agentsmineralizing the

N in the field. Microbial population counts and chemical indices re-

quire a greater amount of extrapolation. By far the greatest amount of

work using this technique has been with agricultural soils. However,

extensive incubation studies of forest soils have been conducted by

Europeans in which raw humus decomposition was involved (Overrein, 1970a;

Overrein, 1970b;Popovic, 1975). Williams (1974) investigated the effect

of water table level on N mineralization in peat from coniferous

forests in England. Tamm and Pettersson (1969) studied N mobilization

on limed and unlimed A and B soil horizons from forest soils in Sweden.

As part of the Solling Project in West Germany, Runge (1971) investi-

gated N mineralization in situ within different soil horizons of several

forest sites using soils contained in polyethylene bags. The same

technique was used by Melillo (1977) in the U.S., but very few other

studies of N mineralization in forest soils using incubation techniques

have been reported.

Research on nitrogen mineralization using incubation techniques

has been motivated by the need for rapid and reliable methods of

assessing N availability in agricultural soils. Stanford and Hanway

(1955) developed a technique that has been adapted for routine use in

the Iowa State College Soil Testing Laboratory. The procedure was based

on the amount of nitrate leached from soil columns after a two week in-

cubation period. Since that time Stanford et al. (1974) have suggested

that the soil nitrogen mineralization potential (NO) be used as a basis

for predicting actual N supplying capacities of soils. NO is considered

to be the quantity of soil organic N that is susceptible to mineraliza-

tion according to first-order kinetics (Smith et al., 1977). Limited

evidence from greenhouse studies involving fluctuating temperatures and

near-optimum soil water contents lends support to this approach

(Stanford et al., 1974). A more recent application of the mineralization

potential approach predicted with some success field N mineralization

for a variety of soils in southern Oklahoma (Smith et al., 1977). Field

N mineralization was measured on a monthly basis using soil in plastic

bags and glass filter tubes.

The development of the N mineralization concept came about in

connection with a study by Stanford and Smith (1972) conducted to assess

the long term mineralization capabilities of soils differing widely in

chemical and physical properties. Estimates of NO were made on the

premise that N mineralization rate under a certain set of environmental

conditions is proportional to the quantity of mineralizable substrate

in the soil. Results of their study show that this assumption is

generally valid because mineralization rate constants associated with

the determined values of NO were similar for a broad range of soils.

The relationships between soil N mineralization and soil water

content and matric suction were later studied by Stanford and Epstein

(1974). Their results showed that in the range from optimum soil water

content (0.1 to 0.3 bar) to 15 bars, a near-linear relation generally

existed between amounts of mineral N accumulated and soil water contents.

Decline in mineralization with decreasing temperature generally

follows an asymptotic curve that approaches zero (Harmsen and Kolenbrander,

1965). Over the temperature range normally encountered in the field,

Stanford et al. (1973) found that a temperature coefficient (Q10) of 2

predicts nitrogen mineralization fairly well over a range of soils. In

a study of soil nitrogen mineralization potentials under modified field

conditions, Smith et al. (1977) concluded that N mineralization can be

estimated from NO when soil temperature and soil water variations are

taken into account by using the above relationships. This general

approach apparently is applicable for arable crops, but its utility for

range and forest conditions is yet to be determined (Stanford, 1978).


Study Area Description and Characterization


Longleaf-slash pine forest communities have developed through a

complex interaction of climate, physiography, soil, and biota (Croker,

1968). The north Florida region has mild winters and hot, humid summers.

It enjoys a long growing season with plentiful rainfall (1370 mm/yr)

and mean annual temperatures ranging from 150 to 270 C. Physiography

of the region is characterized by flat undulating topography with a

multitude of small ponds and swamps. Also common are lakes, sinkholes,

and depressions that are typical of recently emerged limestone regions.

Soils (mostly Spodosols and Ultisols) have generally derived from

marine deposits of sands and are siliceous rather than calcareous and

are usually acid and infertile. Other than pines, vegetation growing

in the region includes many deciduous trees and shrubs, grasses, and

forbs. The pines and grasses are usually shade intolerant but resistant

to fire, whereas the deciduous trees and shrubs are shade tolerant and

more vulnerable to fire (Croker, 1968).

The region, as well as most of the lower coastal plain, was

originally dominated by vast areas of pure stands of longleaf pine

while slash pine was confined to stream courses, pond margins, and

other usually wet, fire protected areas. Heavy timber cutting, land

clearing, road building, and especially fire control allowed the

invasion of thousands of acres by slash pine (Wahlenberg, 1946).

Overcutting, poor regeneration, and shade tolerance have also favored

a major successional trend toward deciduous shrubs and hardwoods.

There is general agreement that the climax forest in this region is

oak-hickory while the pine forests are a fire subclimax (Croker, 1968;

Boosting, 1956).

The study area is located in the University of Florida Austin

Cary Forest. The forest lies approximately 15 km northeast of

Gainesville, Florida.

Amounts and Description of Vegetation

The stand consisted of 20 ha of naturally regenerated 45-year-old

slash pine and longleaf pine. Stand density and stocking were rela-

tively low with 501.9 trees/ha and 16.18 m2/ha of basal area, respec-

tively. Longleaf and slash pine were distributed in approximately equal

numbers on the site, but longleaf pine made up 71% of the basal area of

the stand. Twelve longleaf and twelve slash pine trees were selected

for biomass sampling according to their diameter size class frequencies

in the stand. Tree stand biomass was 96,200 kg/ha. This was comprised

of approximately 3000 kg/ha in needles, 10,100 kg/ha in branches, and

83,100 kg/ha in the stems (Table 26, Appendix A). Details of the

sampling procedures and methods of analyses and complete results are

given by Garbett (1977).

Typical of many naturally regenerated slash-longleaf pine stands,

poor stocking allowed the development of a diverse and vigorous under-

story. The more open areas were occupied predominantly by wiregrass

(Aristida strict Michaux) while dense clumps of gallberry (llex glabra

(L.) Gray) were found in more shaded areas (Figure 1). Saw palmetto

(Serenoa repens (Bartram) Small) was the dominant species of the under-

story plants covering nearly the total area but varying in height and

density. Understory vegetation was sampled by harvesting all material
within twenty 2 m circular plots located randomly on each soil type.

After drying to constant weight at 650 C in a forced draft oven,

samples were ground to pass a 2 mm stainless steel screen and subsampled

for nutrient analysis. Nitrogen concentrations were determined using

the macro-Kjeldahl procedure (Bremner, 1965a). Further analyses were

performed after ashing 2.000 g samples for 5 hours at 500 C, then

dissolving the residue in dilute HCI. Phosphorus values were determined

by the Murphy-Riley ascorbic acid method (Murphy and Riley, 1962), and

K, Na, Ca, Mg, Al, and Fe by atomic absorption spectrophotometry. The

biomass of the understory vegetation averaged 6913 kg/ha. Nutrient

concentrations of the understory tissue were approximately 75% of that

of the needle component of the trees (Tables 13 and 27, Appendix A).

Forest Floor

The forest floor was generally well developed, but its depth and

structure varied considerably due to physiography and stand density.

Litter materials, separated into 01 and 02 fractions, were collected

from twenty-five 0.25 m2 plots randomly located on each of two soil

types within the study area (both Spodosols). The average amount of

biomass making up the forest floor was 37,680 kg/ha which, like the

tree volume, was somewhat lower than comparable sites in the region.

The 01 and 02 horizons were thoroughly characterized and results were


1. Slash pine, palmetto, gallberry.
2. Longleaf pine, slash pine, palmetto, gallberry, wire grass.
3. Longleaf pine, slash pine, gallberry, palmetto, scattered hardwoods
and ferns.
4. Mixed hardwoods, slash pine, wax myrtle, palmetto, gallberry.
5. Cypress dome, mixed hardwoods, gallberry, palmetto.
6. Open area, occasional hardwood, acquatic plants, grasses, sedges.

Figure 1. Vegetation map of the study area.

reported earlier (Huang, 1978). Horizon thickness, density, pH, cation

exchange capacity, total carbon and concentration and total content of

nutrient elements were determined. The only property significantly

different between the two soil types (both Spodosols) present on the

area was the C/N ratio. However all physical and chemical properties

were significantly different between the forest floor horizons within

soil types. The concentrations of N, P, Fe, and Al were greater in

the 02 horizon while the remaining nutrient cation concentrations were

lower as compared to the 01 horizon (Table 29, Appendix A). The pH of

both horizons of both soils was less than 4 indicating that fungi play

an important role as detrital decomposers (Alexander, 1977). The

average cation exchange capacity (30 meq/100 g), although relatively

low compared to humus, contributed significantly to the total exchange

capacity of these soil systems.


The predominant soil types present on the site and considered in

this study were Electra fine sand which is a member of the sandy,

siliceous, hyperthermic family of Arenic Ultic Haplohumods, and Wauchula

fine sand which is a member of the sandy over loamy, siliceous, hyper-

thermic family of Ultic Haplaquods (Soil Survey Staff, 1975). The Electra

series consists of somewhat poorly drained, moderately slowly permeable

soils formed in sandy and loamy marine sediments on slight ridges in

flatwoods areas of the coastal plain. The subsoil and the lower part

of the subsurface is saturated in summer and early fall. Water runs

from the surface slowly on cleared land during this period but there is

seldom any surface runoff from forested land. The Wauchula series is

similar to the Electra but is found on wetter, nearly level sites. The

soil is poorly drained with a water table at depths of less than 25 cm

for months during most years.

These soils are not used for cultivated crops. In some areas

these soils are used for improved pasture and range, but in most areas they

remain in native vegetation and pine plantations. Slash pine site

index for both soils is estimated to be 80 at age 50 years (USDA, 1929).

Important differences between the two soils relative to pine production

are drainage, organic matter content, and depth and development of the

Al, B2h, and B2t horizons. Both the spodic and argillic horizons

perch water during periods of heavy rainfall, but the overall sandy

composition of the soil (> 90% sand) creates a low moisture holding

capacity. Therefore, the site's hydrology is closely tied to rainfall

frequency and soil type. The spodic horizon of the Electra soil begins

at 55 cm compared with 35 cm in the Wauchula. The argillic horizons

begin at 140 cm and 95 cm, respectively (Appendix B).

In order to thoroughly characterize the soils on the study area,

12 replications of 5 cm diameter soil cores were taken to a depth of

90 cm from each of the two soil types. Although the bottom of the B2h

horizon was at a greater depth in the Electra soil, the Wauchula had a

greater concentration of organic matter in the Al horizon. Therefore,

their total organic matter contents were similar, averaging 153 and

157 t/ha, respectively (Figure 2). Cation exchange capacities were

generally low, but it was considerably lower in the Al horizon of the

Electra soil (1.38 meq/100 g) than in the same horizon of the Wauchula

soil (2.91 meq/100 g). Because organic matter content and total

nitrogen correlated well with cation exchange capacity in these soils,


1.0 2.0


Soil horizon depth and organic matter content of the soils
found on the study area.

Figure 2.

their values for these properties have a similar pattern (Table 2).

Interesting aspects about the properties of both soils compared with

other forested Spodosols in the region are low cation exchange capacity,

pH, organic matter, total P, and extractable P. Characteristic of most

Spodosols in the region are high levels of extractable Al in the spodic

horizon. The extractable Al in the surface horizon of these two soils,

however, is relatively low and may be responsible for the low levels

of total P. The low productivity of the stand was due, at least in part,

to these low P levels.

Soil moisture and temperature were monitored continuously in the

stand for part of 1977. The overall mean temperature at a 10 cm depth

for the months of September and October was 26.00 C, ranging from 24.30

to 27.70 C. The mean soil moisture content at 10 cm for the period of

July through December was 9.2% (Bastos, 1978).


Temperature, precipitation, and relative humidity are probably the

most significant climatic factors affecting productivity of pine eco-

systems in this region. Three U.S. Weather Bureau Climatological sta-

tions within the county collect temperature, precipitation, and freeze

data. Hourly wind direction and velocity, temperature, humidity, and

pressure are collected at a CAA meteorological facility at the

Gainesville Airport which is 10 km from the study area. The School of

Forest Resources and Conservation and Soil Science Department of the

University of Florida established a climatic station in the center of

the study area in October, 1975. The station collects data on wind

vector, solar and net radiation, pressure, air and soil temperature,

Table 2. Physical and chemical properties of the two soil types present
on the study area.

Property Electra Wauchula
Al A2 B2h Al A2 B2h

Thickness (cm) 20.8 32.9 36.5 20.5 15.8 33.25

Bulk density (g/cc) 1.53 1.84 1.92 1.67 1.83 1.93

Sand (%) 93.3 93.8 91.68 93.3 94.2 92.1

Silt (%) 5.8 5.2 5.8 4.3 3.5 4.1

Clay (%) 1.0 1.0 2.5 2.4 2.2 3.9

CEC (meq/100 g)/ 1.38 0.22 2.00 2.91 0.96 1.94

pHb/ 4.28 5.06 4.73 4.20 5.00 4.72

Organic matter (%)c/ 1.30 0.29 1.36 1.72 0.50 1.40

Total nitrogen (ppm)d/265 66 219 284 93 245

Total P (ppm)-ef 26.8 16.4 37.3 25.6 19.6 39.0

C/N ratio 29:1 27:1 37:1 34:1 35:1 36:1

Extractable elements
Phosphorus (ppm) 2.68 1.06 1.15 2.04 1.04 0.90

Potassium (ppm) 15.57 2.80 3.73 13.77 2.63 2.53

Calcium (ppm) 59.57 26.70 25.40 62.90 30.10 27.67

Magnesium (ppm) 10.87 3.80 3.50 12.73 4.67 4.00

Sodium (ppm) 38.33 36.67 39.33 36.33 36.33 33.67

Aluminum (ppm) 27.33 22.67 207.00 37.33 27.00 252.00

Iron (ppm) 7.23 4.87 5.27 9.77 7.93 9.90

a-NH4OAc buffered to pH 4.2
h2:1 water-soil suspension

Walkley-Black wet oxidation

d-Macro-Kjeldahl (Bremner, 1965a; Jackson, 1958)

Sodium carbonate fusion (Jackson, 1958)

Double acid extract

relative humidity, evaporation, soil moisture tension, depth to water-

table, and wetfall-dryfall atmospheric deposition. Selected parts of

the data collected at these facilities were used in the study.

Average annual precipitation for the Gainesville, Florida,area was

1370 mm/year (30 year record). The largest proportion of the annual

rainfall occurs during the months of June, July, August, and September.

Two dry periods commonly occur during the year, the first beginning in

March and extending through May and the second during the months of

October and November (Figure 3). Rainfall can be extremely variable

and departures from the mean annual level can be as great as 40%

(Dohrenwend, 1978). Snow is not a significant factor in this region.

The average annual air temperature for the area is 21" C with a range

of 130 C. The average frost free season is 295 days with freezing

temperatures occurring on the average only four times per year. The

average annual soil temperature at the 10 cm depth is 23 C with an

annual variation of 16" C. The warmest temperatures observed at that

depth were in July and the coldest were in February. Average monthly

relative humidity usually stays between 40 and 70% (Dohrenwend, 1978).

Control Stand Installations


The amount and concentration of nutrient elements and other se-

lected cations in the throughfall and stemflow of the stand were deter-

mined for one year from May, 1977,to May, 1978. Collections were made

in a 5 ha uncut portion of the stand. A 40x40 m grid network was laid

out over the area. Line intersections marked with flags were used to





S 4-




i o
0 >0


1 0o


Z 0D -- CD0

<- 0

< o\

(' ) N 0 0 1 1
a r~ A u- u s

>C QD \ Q)%

0 % LA 0 a 0 L\ >0
0 w- C'. 0 3-

locate the precipitation collection devices. Twenty-four through-

fall collectors, 12 for each soil type, were located randomly on grid

intersections. The collectors consisted of 1 liter Nalgene bottles

with 15 cm diameter cylindrical funnels held in place with rubber

stoppers (Figure 4). Nylon screens (1 mm mesh) were placed inside the

funnels to prevent litter from entering the bottle. Three additional

collectors were collocated with Taylor plastic rain gauges on wooden

stands located in the clearcut area. Precipitation from these latter

collectors was used for comparison with the throughfall and stemflow so

as to separate nutrient inputs from precipitation from inputs from the

throughfall and stemflow solutions. The three rain gauges located with

the collectors were used to measure rainfall and calibrate the funnels.

Stem flow collars were attached to 22 trees in the control stand.

The 40x40 m grid was used to randomly locate 5 plot centers on each

soil type. Collars were attached to all trees (with diameters greater

than 7.5 cm) within 30 m2 circular plots. Tree diameters ranged from

7.6 cm to 31.5 cm (Table 3). Collars were constructed from foam rubber,

duct tape, and caulking compound (Figure 4). After the rough outer

bark of the tree was smoothed with a knife, a 3x3 cm strip of foam

rubber was wrapped around the tree at breast height and taped in place.

As the tape was wrapped around the foam rubber, a 3 cm lip was allowed

to extend above the foam strip which formed a 3x3 cm trough around the

tree. This trough was lined with caulking compound to form a flexible,

waterproof stemflow collar. Tygon tubing was fitted to the collar and

run to a series of 3.8 1 plastic milk containers placed inside a lidded

114 1 plastic garbage can.





















Table 3. Layout for stemflow collection: Number and diameter of trees
within ten 30 m2 circular plots on two soil types.


Plot no.

dbh (cm)

Plot no. dbh (cm)










No trees found within the boundary of plot F 2.

Rainfall, throughfall, and stemflow were collected on a rainfall

event basis. After the sample volumes were measured, 100 ml subsamples

were stored for up to 2 weeks at 4 C. A 100 ml combined sample was

composed of samples collected within a 2 week period. Samples were

combined in proportion to the amount of rainfall each represented.

Chemical analyses performed on these water samples included pH, NH4-N,

NO -N, total P, K, Na, Ca, Mg, Al, and Fe. NH4-N and NO3-N were deter-

mined by micro-Kjeldahl distillation (Bremner, 1965b).


Litterfall was collected at monthly intervals from May, 1977, to

May, 1978. Eighteen 1 meter square frames were built from 2x10 cm lumber

One millimeter mesh nylon screening was stretched across the bottom of

the frames. Wooden legs were attached to the frames to hold traps a

few cm off the forest floor (Figure 4). Grid intersections were used

to randomly locate nine traps on each soil type. The monthly litter

collections were dried to constant weight at 650 C in a forced draft

oven, weighed, ground in a Wiley mill through a 1 mm stainless steel

screen, and analyzed for total N, P, K, Ca, Mg, Fe, Na, and Al. Samples

were analyzed by the same methods outlined for the understory


Harvesting, Site Preparation, and Planting

After the trees, understory, forest floor, and soil were sampled

and characterized, the stand was harvested during August, 1976. An

adjacent 5 ha stand was left as a control. Conventional clearcut

methods were used. Trees were felled with a chainsaw, delimbed with a

delimbing gate, topped at a 6 cm diameter, and dragged to a central

location with a skidder. After the harvest was completed, the remain-

ing non-merchantable pine and hardwood trees were felled and logging

slash was distributed evenly on the site with a D-6 Caterpillar tractor.

During October, 1976, the clearcut was broadcast-burned except for

four 40x40 m plots. During November, two intensities of mechanical

site preparation were applied in factorial combination (2 factorial)

to burned and unburned plots on the two soil types (Figure 5). Half



I ..*





Figure 5. Soil map and experimental layout of the study area.




the site was prepared by double chopping with a drum chopper and half

was prepared by first blading all logging debris, understory, and

forest floor into windows, discing twice with a double-gang disc

harrow, and bedding the area in 3 m rows.

The following spring, during March, 1977, the study area was

planted with unimproved 1-0 slash pine seedling stock obtained from

Container Corporation nurseries located at Archer, Florida. The

intensively prepared areas were machine planted in the beds at 3 m

between row and 1.5 m within row spacings. A large amount of logging

debris on the chopped areas prevented the use of a machine planter,

therefore seedlings were hand planted at the same spacings as machine-

planted seedlings. A five-week drought followed the planting operation

which immediately resulted in an estimated 50% mortality on all

preparation treatments. Within four weeks of the original planting

date, most of the dead seedlings were replaced. Because the within

row spacing was unusually close, a reasonable stand (approximately

1000 trees/ha) was established and, in some areas, may have to be

thinned before the trees become competitive.

Harvested Area Installations

Soil Solution and Water Table

Soil solutions from both the harvested and control areas were

collected biweekly for the first year after harvest, followed by

monthly samplings during the second and third years. Soil solution

samplers consisted of 4 cm diameter sections of PVC tubing with porous

ceramic cups cemented to one end. A two-holed rubber stopper (no. 10)

closed the opposite end but allowed the evacuation of air and the

pumping of soil solution from the tube. The tubes were installed in

the field in sets of four tubes of different lengths. When installed,

the ceramic cups of each set were at 15, 30, 60, and 90 cm depths which

corresponded roughly with the lower parts of the Al, A2, and B2h

horizons and immediately above the argillic horizon, respectively.

Three sets of tubes were installed in each of six treatment combinations

on the clear-cut, and on each of the two soils in the control area, for

a total of 30 sets or 120 tubes. Soil solution samples were analyzed

for the same elements and by the same methods as the precipitation

solutions described above.

Water table level monitoring tubes, consisting of 4 cm diameter

PVC tubing, were collocated with each set of soil solution sampling

tubes and were installed to a 180 cm depth. Water table level was

measured at the time soil solution was sampled.

Soil Sampling

In order to characterize the soil prior to harvest and site

preparation, 12 replicate 5 cm diameter soil cores were taken from the

surface to a depth of 90 cm from each soil type. Bulk density was

determined gravimetrically using an oven set at standard 1050 C. Sand,

silt, and clay fractions were determined using the hydrometer method

of particle fractionation (Day, 1965). Soil pH was measured in a 1:2

soil-water suspension, and cation exchange capacity was

estimated using NH40Ac buffered to pH 4.2, the average pH of the soil.

Soil organic matter content was determined using Walkley-Black wet

oxidation procedures, and total N by standard macro-Kjeldahl procedures

(Bremner, 1965a). The remaining elements were extracted with a double

acid extract (0.025 N H2S04 0.05N HC1) that had: been correlated ,with

plant available nutrients and is widely used in many southern states.

The soil extracts were analyzed by the same methods outlined for the

soil solution and precipitation samples.

In an attempt to monitor changes with time in soil physical and

chemical properties, the soil was sampled at the end of each of the

first two growing seasons. After the first growing season, three

replicate soil samples were taken from each of the treatment areas

during the late summer (August). Samples consisted of soil cores 15 cm

diameter by 20 cm deep. The sample included the litter layer, if any,

and all other debris that may have been mixed into the surface soil as

a result of site preparation. Due to the looseness of the recently

prepared soils, bulk density was determined by displacement. The

volume of the soil sample was determined by lining the cylindrical hole,

created when the sample was extracted, with a thin plastic bag and

filling it with water level with the soil surface. The volume of water

used to fill the hole was assumed to be the volume of the soil extracted.

In order to obtain a more representative sample during the second

growing season, each soil sample was made up of 15x20 cm subsamples.

Using the soil solution sampler installations as the center, subsamples

were taken at the cardinal and intermediate points around two concentric

circles with radii of 7.5 and 15 m. Subsamples were dumped into a

portable cement mixer, and, after thorough mixing, a composite sample

was withdrawn. Because the soil surface was more stable the second

season, a 5.95x5.35 cm Uland-type soil coring device was used to sample

the soil for bulk density.

The soil samples collected during both growing seasons were

analyzed in the same manner. They were first sieved through a 2 mm

sieve, effectively separating the soil from the litter and other

organic debris. This organic fraction was dried to a constant weight

at 650 C and weighed. Soil organic matter content was determined by

loss on ignition (Jackson, 1958). Cation exchange capacity, pH, total

N, and extractable P, K, Ca, Mg, Fe, Na, and Al were determined by

the same methods used for the initial soil characterization. Soluble

carbon, determined only on unsieved subsamples taken at the end of

the second growing season, was analyzed with a Beckman total organic

carbon analyzer from a 1:10 soil-water extract by combustion oxidation

of the carbon and subsequent infrared analysis of the evolved CO2.

Photo Points

Permanent photo points were located in each treatment area to

monitor the growth of the pine and its competing vegetation. Points

were marked with a 1 m high section of PVC tubing over which a camera

tripod was placed. After leveling the camera with a pen level and

aligning the view finder on a range pole, repeated photos of precisely

the same area could be taken. A quarterly photo sequence of each

treatment area was taken for the first two years.

Competition Evaluation

Treatment effects on biomass and nutrient content of the competing

vegetation were determined. Using the soil solution tube installations

as a center, four 1 m2 subplots located in the four cardinal directions

3 m from the center, were harvested. Vegetation was separated into

woody, forb, and grass-types plants. Vegetation of each type from

each of the four subplots was combined for three replicate composite

samples from each treatment. The vegetation was dried, weighed,

ground, and analyzed using the same methods outlined for the litterfall.

Tree Measurements

Pine seedling survival and growth were measured on all treatment

areas. Survival, height, and tipmoth infestation were measured in

January after the first growing season. Survival, height, and diameter

were measured in January after the second growing season. All the

trees in the four 40x40 m unburned plots were measured while each of

the four burned-plot treatment areas was sampled by measuring three

20x40 m plots. A meter stick and calipers were used to make the

measurements. Needle tissue was taken from each tree within alternated

rows. There were 13 rows within the 40 m width of a plot so six rows

were sampled. The tissue was analyzed for total N, P, K, Ca, Mg, Fe,

Na, and Al by the methods described for litterfall and competing


Laboratory Mineralization Study

The nitrogen mineralization potential of soils from each treatment

area and the control area was determined using a modified version of

Stanford and Smith's (1972) incubation technique. The 30 composite

soil samples collected from the field during August of the second

growing season were used in this experiment. Within a week of the

sampling date, and before they had dried, the soils were sieved through

a 2 mm sieve to remove the litter, large roots, and other organic

debris. Although the coarse organic materials were removed, the humus

and a large fraction of the fermenting layer were included in the soil

used in this experiment.

Four replicate subsamples of approximately 250 g of soil

(corrected for moisture) were limed with 1.75 meq of Ca(OH)2. The

soil and reagent grade Ca(OH)2 powder were mixed by pouring both into

500 ml Erlenmeyer flasks which were then stoppered and shaken for

10 minutes on a wrist action shaker. Four x20 cm PVC incubation tubes

were prepared by stoppering one end with a one-hole rubber stopper

(no. 10), then packing a 2 cm thick plug of glass wool into the tube,

followed by a 54 mm diameter glass filter. The limed soil was then

transferred to the incubation tubes and followed by a second 1 cm thick

plug of glass wool (Figure 6). The soil was poured into the tubes at very

nearly field density (approximately 1.5 g/cc). The soil cores were

moistened by pouring 25 ml of distilled water into the tops of the

tubes. After remaining for five days at room temperature, the soil

cores were leached with 250 ml of 0.01 M CaCl2 solution followed by

100 ml of a Hoagland minus -N nutrient solution (Hoagland and Arnon,

1950; Hewitt, 1966). Leached of their initial NH4-N and NO 3-N,

reconditioned nutritionally, and equilibrated at optimum and uniform

moisture content, the soil cores were placed in an incubator set at

280 C. The soil cores were leached every 25 days for the following

125 days.

When leaching, the CaCl2 solution was allowed to saturate the soil

core under gravity. Within 2 min the cores would begin to leach and

exactly 2 min from the time the CaCl2 solution was applied, a vacuum

pump evacuated the soil cores at 0.065 atmosphere of suction for an






Illustration of a soil core used in the N mineral-
ization study.

1 cm

20 cm
20 cm

2 cm

Figure 6.

additional 10 min. Leachates were collected in 250 ml plastic centri-

fuge bottles and weighed to determine their volume. The 100 ml of

minus -N nutrient solution was applied and leached through the cores

in exactly the same manner. This solution consisted of 0.00025 M

Ca(H2PO4) H20, 0.0005 M KH2P04, 0.002 M MgSO4*7H20, and 0.0015 M K2SO4

providing P, K, and Mg in the same proportions as the solution used by

Stanford and Smith (1972). In this study, however, KH2PO4 was used in

combination with Ca(H2PO4)2.H20 to provide a more highly buffered,

pH 5 solution.

A leaching apparatus (Figure 7) was specially designed with

features that allowed expeditious and carefully timed saturation and

leaching of the soil cores, uniform application of solutions, and

easily controlled and uniformly applied leaching tension. Leachate

pH and NH4-N were determined. Micro-Kjeldahl distillation was used

for the N analyses (Bremner, 1965b).

Figure 7. Soil core leaching apparatus used in the N mineralization


Site Characterization

Biomass Summary

The effects of soil series and tree species on stand biomass and

nutrient concentration, content, and distribution were determined by

Garbett (1977). Summaries of stocking, biomass and nutrient distribu-

tion, and nutrient concentrations by tree component are provided in

Tables 26, 27, and 28, Appendix A.

Due to the greater amount of understory biomass on the Electra

soil, total nutrient contents for all elements, except Na, were higher

on this soil than on the Wauchula (Table 28). Nitrogen was the only

element significantly different in concentration between the under-

stories on the two soils. It was higher in the vegetation found on the

Wauchula soil, perhaps as a result of different plant species or a higher

soil N content. Although the understory made up only 7% of the total

above-ground biomass, it contained from 20 to 30% of the total quantity

of each of the measured elements, except Al. The trees held 94% of the

Al found in the total biomass.

The amount of organic material, nutrient concentration and content,

and physical and chemical properties of the 01 and 02 forest floor

horizons were determined by Huang (1978). Summaries of biomass and

selected element concentrations and contents are provided in Tables 29

and 32, Appendix A.

Soil physical and chemical properties for the Al, A2, and B2h

horizons were determined on soil samples taken immediately prior to

harvest (Table 2). An analysis of variance, designed with effects of

soil type split by effects of soil horizons (Table 31, Appendix A),

was used to evaluate the soil properties as influenced by these

factors. Soil depth, percent silt, percent clay, and amounts of ex-

tractable P, Na, Mg, and Fe were all significantly different between

soils. There were significant differences in the values of all soil

properties, except for Na, among horizons. Detailed results of the

analysis are provided in Table 33 of Appendix A.

The biomass and chemical element contents of the vegetation,

forest floor, and soil components of the ecosystem are summarized in

Table 4. Comparisons of the tree and forest floor biomass and nutrient

content show that there is three times as much biomass and K in the

tree component than in the forest floor. Phosphorus, Ca, and Mg were

approximately evenly distributed, N was twice as high, Al was four

times higher, and Fe 55 times higher in the forest floor than in the

tree component.

As might be expected, the soil contains the greatest amount of all

the above mentioned elements. Comparisons of quantities of organic

matter, total N, and the remaining extractable elements in the soil

with the biomass and total element content of the sum of the remaining

components show that there was about an equal amount of organic material,

six times as much N and Fe, 0.8 the amount of P, about 1.5 times as much

K, Ca, and Mg, and 30 times as much Al in the soil than in the sum of

the remaining components.















U 4-


th c

E r




c c L.
-D C 0 4-' 4-'
N f0 (- 0
Sa)- 0 5
0 4
I-. I--

* : **- 7 o o

-- T

o Lt CO CLAO 0
o- C co lN

C4 C- L-\ 0

- M uA -

- MN o N <1

0 0 u ON

0 -- LA rlct1

Y. Lo LA


CO-30 0
CM r- 04 CM4

r- cr -T -

0 -
f CO\

00 -


m er-, 0 cCoO

M : r3- %o c

o cO r o 0r r 'O
L"% 0 LA 00 CO M en

M- M

n-o ooor

CMf C i

} C
O- L



0 4-
N a)
L- L
0 0
.C 4-

0 10

0 0

-c N

0 -
N -
C (U

-'.0 nio
- *-9 *0
. .
S00 .r

0 N IL

-1 r C





a) 4ca

c X

3 a

I .Q
> 0
0 -a

0 1I
L1 .


0 0 0


-0 c -
4-' O(
a (U -

0 0 3

0 (

1. 4-' L ^:

(Cl C' f/ 0..






(u 0



-" 0

E 4-


Precipitation and Litterfall Inputs

During the year beginning May 18, 1977, precipitation for the re-

search area totaled 1425 mm. Throughfall was 1396 mm or 96% of total

precipitation, which meant that only 4% of the total precipitation was

intercepted and evaporated by the forest canopy. Stemflow amounted to

4.4% of the total precipitation volume. The concentrations of all

measured elements in the stemflow were significantly greater than in

precipitation and throughfall, except that the concentration of NO -N

did not differ among these sources. Magnesium was the only element with

a significantly greater concentration in the throughfall than in

precipitation (Table 5).

Mineral input in throughfall (T) was estimated by

T = P + V (1)
m m m
where P is precipitation and V is vegetation leaching. Similarly,

mineral input in stemflow (S) was estimated by

S = P + V (2)
m m m
No attempt was made to separate dry-fall from precipitation inputs;

however, it was understood that P input was a combination of wet- and

dry-fall. Annual mineral inputs in throughfall were several orders of

magnitude greater than in stemflow for most measured elements (Table 6).

Total solution mineral input (TSm) was estimated by

TS = T+ S (3)
m m m
and vegetation leaching (V ) was determined by

V = TS P (4)
m m m
Values of Vm for NO3-N, K, Mg, Al, and Na were positive which indicated

a net leaching of these elements from the trees. NH4-N, P, and Ca were

negative, however, which indicated that these elements appeared to be

absorbed from the precipitation by the trees.






I 0


I f U




I 0


I (0
I c


I 4
I -






to .0


C 0

"1 4-
0 E
-C 0

I- -/)









9 9


o o

(0 .0

m 0
o o

o o


o 0

(0 -

Cf .0

O -
o o
t0 0(y




















- (
O *-,


















o -

C -
0 Co


4 L

C 0

o- L

L. C

S 1-

O 0
4- -1 +4

+ o
o- C

0 0
4- .4-'
-C 0 3

3 0

*- --

II O *-
4- II 4
3 01 1 (D

c in 0
U 1

o ) c 0

o 0 o

4- O- 0

h- > I-
o :o ^










E c 0 c C
I- 0 L 4
E C 0 0.
- U 0 0 C
4-' C j .
E 4-' 0 0
.C 0 4-- *-
m0 -- O. 0. to 0o
3 4- 3: (4 4 i)
0 E OU 0 J -
a E- 0
-C 41i m V o
- t0 -- (n 4> 4J 4-


Annual litterfall totaled 5444 kg/ha with the largest amounts

falling during October through December (Figure 8). Mineral concentra-

tion in the litterfall varied considerably during the year. Nitrogen,

P, and K concentrations were lowest during peak litterfall periods and

highest during March, a period when litterfall was minimal (Table 7).

Annual mineral returns in the litterfall (Lm) (Table 8), compared to

total solution inputs, were about six times greater for N, four times

greater for P, nearly the same for Ca, Fe, and Mg, but less than half

for K and only one-fourth for Na (Table 6). This showed that total

solution input is an important source for several essential nutrient


Total mineral input to the soil (TMm) was estimated by the sum of

solution and litterfall inputs:

TM = TS + L. (5)
m m m
Ratios of N, P, K, Ca, and Mg in total inputs were 18, 1, 6, 20, and 6.

S- o o o0


m' O



L4 0

4 -J






(eq/5g) SSVWOi9 sL

co \s .-
(N L O o CM

--T Ln -_; -T

rM\ 0-0
-0 m W

a 0 mn Mo LAo M
O r Ln "M -:
r- Lfn M n 0co o

mN mM C N LAn -r
N- \o m rN C
mt mn -:r --T -:r -r Ln

Lrn a) m ~ N- LA r-
0-% 1o *lo o a) ai 0o

M 0m m l \DN \M 0 c% 00
- Ci N-roo CC) -3w N -
r^coc o co o-n N-

\0 -- LA -,0
oo -T o u
mtr V MN

M 0 N-
lA r- mo P-
m mt C" CN

U (N OM -3 0 % 00 'O
LA A'\ %D LA Lr\ ft LA' %D

Sc \D 0
O' E -3i \\

rN- o N- m -3T L o o N- LA oo
CN -- %o 4 n -T Ln CN co -T
(N (N (N N (N '- (N (N ft (N C

'\ rN- m rN oo o m r-
'o m O LAoc o cn cN N -
ut (N oL Ln A LA\ N-\ '
-w -w -r -T -T e -w -T

o rw \o
ft r--. f
u ...1-

SN o N-

r oo o o0 (c -, c A I

- LA
- 0o
N- o
CM ft


0 -



o o


m N
-0 o

o o

o o


























Co o o-


0 N( CM

r- n '

n 00- -

0 0 o 0 0 o


o c



o o

N.0 -


m -.


M -a rU '%0

_- (


0 -0 U0 0


"- "- -T Co Co


0 %0 0 -

- 0 ('4
C--c C- CA CA
o c o (N 4


MN O 00 Co N N O I -
- 0 rN -3- CM -3
(N -3- 4c N -M '.01 -3'

I I I I I I I I I I I Cu
fn 0 N- LA Co C\ 'O N- 0i oo U0\ 4-I
-- M CC -0

o C4M o o-o

Effects of Harvesting and Site Preparation

Soil Property Changes Over Time

The effects of site preparation on soil physical and chemical

properties were evaluated each year for the first two years after

harvest. Soil properties included in the evaluation were bulk density,

organic matter content, C/N ratio, pH, cation exchange capacity, total

N, and extractable P, K, Ca, Mg, Fe, Al, and Na. The effects of soil

type and site treatment were analyzed using a factorial design

split across time (Table 34, Appendix A). Because the effect of

burning appeared to be short lived (less than one year), it was not

included as a separate treatment in the analysis. Total N, Ca, and Mg

were significantly different between soil types; however, the major

differences in soil properties occurred as a result of site treatment

and time (Table 9). Organic matter, total N, Ca, and Mg concentrations

decreased significantly with treatment intensity. Calcium, Mg, Fe, and

Al increased, and Na, P, and C/N ratio decreased with time. Total N

appeared to increase after the first year but decreased again by the

second year. Site preparation intensity had no effect on bulk density,

but after the first year there was an overall significant decrease from

the initial value. This improbable result could be due to sampling

technique because it was necessary to use a different sampling method

each year.

An attempt was made to evaluate treatment effects on the amount of

organic matter and total N remaining in the surface 20 cm two years

after site preparation. The control stand was assumed to represent the

condition of the clear-cut area before it was harvested. An analysis of

LA m LA0

e M -

0 C C4
cr 00

00 mo-

- CM C1M

en LO '.0

Co4 -T M
- r^>0 L

cN -*-- f
CM (M f

*- CM


Co co

-L 00 C

ft -
IA oo r

SO (3"
* *








o r




oC o
- 04C -


ci ru


m. M

CN0 0-

- CM4

( .4
M fL f-
' .-- -30

o r. LA



. 1.. L. L

-~0 0 >- >- >-

z z
z 4:

z z z

()1 -I :
Z -i -K



Z z

*it Z Z

0 L- LC
a 0U U 0
Q.>- > >-











variance followed by a Duncan's Multiple Range test showed that the

total content of both properties was significantly different among

treatments (Table 10). The organic matter and total N contents of the

> 2 mm organic fraction were not significantly affected by site prepara-

tion treatment, but site preparation at both intensities caused

reductions in levels considerably below that of the control. Organic

matter and total N contents of the fraction passing through a 2 mm

sieve, on the other hand, were not different between the control and

the less intensive site treatment area, but the intensively prepared

area contained significantly less than either the control or the less

intensive treatment area. Soluble carbon content decreased signifi-

cantly with increasing treatment intensity, butnot in proportion to the

decrease in organic matter. The total organic matter-soluble carbon

ratios for the control, less-intensive, and intensive site treatment

areas were 81, 108, and 138, respectively.

Soil Moisture and Temperature Changes

The effect of harvest and site preparation on soil temperature and

moisture was evaluated by Bastos (1978). A summary table showing treat-

ment and depth effects is provided (Table 35, Appendix A). The highest

daily temperature maximums at the 0, 10, and 20 cm depths occurred on

the intensively prepared area. Daily temperature maximums increased and

temperature minimums decreased with treatment intensity at the 0, 10,

and 20 cm depths. Mean daily temperatures of the soil in the control

area appeared to be about 30 C greater than either of the two treatment

areas of the clear-cut. Below 20 cm, site treatment had little effect

on soil temperature.













0 *

r- 0





0 -




.0 .0
mA cr
Lt1 N

- 0



1 0.

0) 0



c- -

3 0



0 0
0) -o



0 V





o o
o o
0 0o

I -0
0 C0
o o
o 0

No -
oD .0

- 0\D

- 0

h0 (0


.a 0

0 m
V) 0

*- L-

M 4J

L- to























Average soil moisture content increased with harvest and site

preparation intensity at all four of the monitored profile depths. The

average soil moisture content was greatest at 50 cm followed, in

descending order, by the 5, 10, and 20 cm depths.

Water Table Levels

Water table level was monitored from the time of stand establish-

ment through the summer of the second growing season. During this time,

the average depths to the water table in the control stand and in the

chopped and intensively prepared clear-cut areas were 78, 69, and 62 cm,

respectively. Clear-cut harvesting caused a significant reduction in

the depth to the water table. The depth to water table appeared to de-

crease with increased site preparation intensity (Figure 9), but the

difference was not significant. Due to an unusually dry period during

April and May, 1977, the water table dropped below the level of the

piezometers. It did not recover to a measurable stage until the follow-

ing September, with the beginnings of heavy rains. During the same

period, the water table in the clear-cut area remained within 120 cm of

the surface. During February, March, and April of the second growing

season, no difference in depth to the water table was detected among

treatment areas. Water table level responded relatively quickly to

rainfall inputs (Figure 9).

Nutrient Mobilization

The effect of harvest and site preparation intensity on the concen-

trations of selected elements in the soil solution was monitored over

two growing seasons. The effects of soil type, site preparation treat-

ment, burning, soil depth, and growing season were analyzed using one of

0- .------...
-------- INTENSIVE

J A S 0



TIME (months)

Figure 9. Effects of harvesting and site preparation on water
table depth.






7- 1 T.1- -A-7

two statistical models. Soil type and site treatment effects, which

included the control stand and two intensities of site preparation on a

clear-cut area, were analyzed using a factorial design split by depth

at four levels and season at two levels (Table 36, Appendix A).

Because burning was a factor only on the clear-cut area, it was analyzed

by factorial design with soil type and site preparation treatments each

at two levels. The 23 factorial design was split by two seasons

(Table 37, Appendix A).

Except for Ca, Mg, and Na, the solution of the Electra soil had

higher concentrations of the measured elements than the Wauchula soil

(Table 11). Concentrations of NO3-N, K, Fe, and Al were all signifi-

cantly greater. Soil pH and the concentrations of Mg and Na were signi-

ficantly greater in the soil solution of the Wauchula soil. Intensive

site preparation appeared to cause an increase in the concentration of

every element measured in the soil solution. Significant increases

were determined for NH4-N, P, K, Mg, Fe, Al, and Na. Harvesting and

chopping alone had little effect on nutrient concentrations. Soil

solution pH remained relatively constant among treatments at about pH

4.6. There was a significant interaction between soil type and site

treatment factors for soil solution pH and K concentration.

Figures 10, 11, 12, and 13 show how pH and concentrations of

NH -N, NO3-N, and P fluctuated with time and site treatment during the

first growing season. Rainfall patterns appeared to correlate with

many of the fluctuations. Measurements were begun at planting time

which was approximately three months after the clear-cut area was pre-

pared. For the first eight months of the first growing season

I (0 -Q (U (U .0
-O CO O 0 N- I
0) JO I o o N- I

4- z I
4"1 Z I f
U I coo ro


O I ICM Or o --r I ()
< I I
In I I

o -I I

" I o.0 I
4 0) LiA LA LA \ D0 I (0
C. I -
) I 0 0 000 I 0 U
U I L >
C I I 0. -
O I j 4.'
U I I 4- U
01 .0 (0 U I0 O U
" I N- Lt N- I Q.
c1 O.I r 0 I 0 -
(U 2 z Ua

I O- oC > .

C (0 CL0 o( tD L A 1 C

--1 CJ c o o en U) ( 4
O .0 (U (U 0 4 4

0 oo mo o cA

I- ,*- D-. ,. xo .. .
4o1 .0 U 4.

c I ** C LnM ) )
( U I( -U O -0oN 1 O

S0 1 04 00 L I U U
I I 4- ac
S 3 I ( 0 m I -

C I C!- -,-
( I 0 0 0 0 I *J-

0 Z (U .0 (U (U (U 0 C
0 0 I 4- I Z

4 I 0 0 0 0 I 4-
W (.I I a

I I o1 cO r-. o 1 U) U

C 4 I m (U
10 I 0 0 000 I U
SI (U *
0 I I O .-
0. I I
I ( .0 (U f (0 I (A C
41 I 0. I o N- -a 0 : *
I o I *

ZI 0 a 0 0- 1-0 .
4) 0- 4.

i- (D (D U m *c
SU C- .C 4- 0.c
e c .0 z U

C 0 0- 0 0 "0 E t

CL : 0 L) O nE
( U.( U 0 4.' (0 0 (UC

0 --
u- c. o -











11111 I~lI 4iLI~r


TIME (months)

Figure 10. Effects of harvesting and site preparation on soil
solution pH.













< 10.0

< 7.5

2. 5


TIME (months)

Figure 11. Effects of harvesting and site preparation on soil
solution NH4-N concentration.

9 '
* 9
* I
I '
* S
* I
* I
* I
* I

i 1 111111111111




1 ii ll,

0 N D J F

L iUit~


TIME (months)

Figure 12. Effects of harvesting and site preparation on soil
solution NO3-N concentration.








1 1 i1 1

1.1 1111111111111-



0 Z
C. 0 I-
lz o

IO _
0 T- 0

,,o m, .-3 3 ,,

I ( 0 -
I 0

ILL. 1

I In

4I 4-

I' E <.

-- -.

I d)U-

--0 0 0

L -N ,

^____v .

soil solution pH was higher on the intensively prepared site than on the

chopped site. The soil solution pH of the control area fluctuated con-

siderably, but as the site became drier in the Fall, pH values for all

three areas converged at about pH 5.0 and remained relatively the same

until substantial rainfall increased the soil moisture levels again

during the winter months at which time the acidity of all soil solutions

increased (Figure 10).

Ammonium-N concentrations fluctuated considerably among the three

areas, but the highest concentration at any one time was usually found

in the soil solution of the intensively prepared area (Figure 11). The

concentration was highest in all three areas from September through

December. This time spanned both wet and dry periods so there was no

apparent correlation between rainfall and NH4-N concentration in the

soil solution. Concentrations of NO -N were several times greater than

NH4-N, especially during the Spring and Summer (Figure 12). The

levels were considerably greater in the clear-cut areas during this time.

At one point during June, the NO -N concentration reached 1.6 ppm.

Beginning in October the NO -N concentrations decreased abruptly and

remained at a very low level through the Fall and Winter. There was

virtually no NO3-N detected in the soil solution of the control stand.

The P concentration in the soil solution fluctuated with time some-

what like NO -N concentration, but it seemed that only the intensive

site preparation (bedding) treatment was affected (Figure 13). Harvest-

ing and chopping appeared to have little effect on soil solution P con-

centrations. There was an unusually high peak during September when the

P concentration reached 5.6 ppm in the soil solution of the intensively

prepared area. This high concentration was short-lived, but it remained

significantly higher than in the control and less intensively prepared

plots throughout the Fall and Winter.

Broadcast burning of the clear-cut slash appeared to have little

long term effect on the concentration of elements in the soil solution.

Only Na was significantly affected and its concentration was lowered as

a result (Table 12). Soil solution pH, however, was significantly

higher as a result of burning. Results of measurements during the first

year after planting show that soil solution pH was higher for most of

the year in burned portions of both the chopped area and the intensively

prepared areas of the clear-cut (Figures 14 and 15). Because of the

sensitivity of the nitrification process to pH, the correlation of NO3-N

concentration with pH was evaluated for these treatments. During that

period of the year when NO -N concentration was highest overall for both

site preparation intensities, it was considerably higher in the burned

portions of these areas where the pH was relatively high (Figures 16

and 17).

Patterns of nutrient concentration with depth varied among elements.

Phosphorus, Ca, Mg, and both forms of N generally decreased with depth;

K and Na, increased; and Fe and Al had the highest concentrations at

90 cm with intermediate levels at 15 cm. Soil solution pH increased with

depth to 60 cm, but decreased slightly from 60 to 90 cm (Table 13).

The effects of time, in terms of growing seasons, on soil solution

pH and element concentrations of the clear-cut area were evaluated for

the periods April, 1977, to April, 1978, and April, 1978,to January, 1979.

Soil solution pH and all element concentrations except Ca and Al de-

creased after the first growing season. Soil solution pH, and concen-

trations of P, Mg, and Na decreased significantly with time (Table 14).

Lo .(
I m -
(0 I I (I)
z I en z

4 (0 (0 *

0 00
I C .- I / t s.- >

In I z z *.
S- -
I I ,,0
I cI ( n

1 0 0 )-

0)O m I- O
S I L %I V 4V) V) L >
SLL. I I Z Z a

C I I 4 U

> I u
SI ( I n W

I If I A
S-- O CO I 0 0
0 (o I I 3t in > L..
S I c

S"- I- -
4-' I .01 o'e C

0 I zo LA U
0 I (E ( 4 0+

I CO 0 u 0 -
0 (U I 4 0 0 4-

-) Z -Z 4 -

4- (- Ul
-o 5 (U 4- o
r 0 (0 0 4
L -O 40 4-

0 I I L1 0 U / ( / 0 0c

C hC I *) *) 2iZe

0 I I C

0 o0 U
q- Z z 1 U t l
4T I nn .- C
3I C 0 ( U

O I I 2 4 u
0 0 I 1 0
t I I I -

"0 a-
m a
0- Z I I & O

.0 T I 0 0 I I/ 4-
- 0 a I* I Z Z 4-

O a 0 0 a

u I 0L
C) a an
a- a *- 0a

4- U I -
0 c I Co 8
1 I M / C
a I Q Z (U

o I I *

U) I I C D U

C I .0 IC
.0 (U

-- C *-
W I 5 I. L L c

4- C C






J A S 0 N D J

TIME (months)

Figure 14.

Effects of burning, on plots that were clear-cut and
chopped, on soil solution pH.










II /
I /


, ,iii

Figure 15.


TIME (months)

Effects of burning, on plots that were clear-cut, bladed,
disced, and bedded, on soil solution pH.

- 4.







E 0.5

z 0.4





2 5.0

Figure 16


TIME (months)

Effects of burning, on plots that were clear-cut and
chopped, on soil solution NO -N concentration.