SOIL ORGANIC CARBON DYNAMICS IN TALLGRASS PRAIRIE

LAND MANAGEMENT

THESIS

Presented in Partial Fulfillment of the Requirements for

The Degree of Master of Science in the Graduate

School of the Ohio State University

By

Joshua Beniston, B.S.

Graduate Program in Envrironment and Natural Resources

The Ohio State University

2009

Thesis Committee:

Professor Rattan Lal, Advisor

Professor Frank Calhoun

Professor Martin Shipitalo ii ABSTRACT

This study was composed of two research components that examined the effects of tallgrass prairie land use changes on soil organic C (SOC). The first study examined changes in SOC and a suite of soil quality parameters in former agricultural soils now under restored tallgrass prairie. This study analyzed soils in a restored tallgrass prairie landscape with the primary objectives of: 1) To assess the changes in soil organic C

(SOC) in restored prairie landscapes and to compare them with the SOC in conventional agricultural systems on the same soil, and 2) To quantify the impact of the restored prairie plantings on soil quality parameters and compare it with soils under conventional agricultural land uses. The primary study site was the Prairie Nature Center at the Marion

Campus of The Ohio State University. The Prairie Nature Center (PNC) is a 4.5 ha site that began with the creation/restoration of tallgrass prairie plantings in 1977. Soil samples were taken from 31year (P77), 13 year (P95), and 8 year old (P00) restored prairie plantings, as well as from lawn (LA) and an annually cultivated corn (Zea mays)/soybean (Glycine max) field (AG) on similar soils. Soil samples were taken from

0-10, 10-20, 20-30, and 30-40cm depth, from 4 sites in each treatment during the summer of 2008.

ii These soils demonstrated significant increases in SOC concentration, particulate organic matter C (POM-C), water stable aggregation (%WSA), aggregate mean weight

diameter (MWD), total porosity (ft), and available water capacity (AWC), and significant decreases in soil bulk density (ρb) associated with time under tallgrass prairie. The accrual of SOC, in the previously cultivated soil, was greatest in the surface layer (0-

10cm) of the P77 soil (3.45%), compared with the AG soil (2.14%). Soil aggregate properties showed significant increases in the P00 (95.48%WSA) from the AG treatment

(71.11%WSA). This indicates that macro-aggregate properties may be restored relatively quickly (<10yrs) under restored tallgrass prairie. The majority of significant changes observed in this study occurred in the soil surface (0-10cm) layer, which may be due to the high proportion of prairie root biomass in soil surface layers.

The second research component observed long and short-term effects of the conversion of remnant tallgrass prairies to wheat production, in north central Kansas. The long-term study worked to describe and quantify the effects the conversion of a perennial plant community to an annual plant-based agricultural system on SOC pools in sites converted from tallgrass prairie to annual agriculture. The long term effects of land use change on SOC pools were analyzed by sampling five farms that contain both annually harvested tallgrass prairie remnants (PM) and conventionally farmed wheat (Triticum aestivum) fields (AG) on the same soil types.

Soil core samples were collected to a depth of 1m in May and June of 2008.

Management effects on SOC pools were assessed by analyzing total soil organic C

(SOC), total soil nitrogen (TSN), microbial biomass C (MBC) and a particle size fractionation of SOC in coarse sand (>250µm), fine sand (250-53µm), silt (53-2µm), and iii clay (<2µm) sized fractions. PM soils showed statistically higher levels of all parameters measured to a depth of 60cm. SOC pools were decreased by 30% in the AG soils (59 Mg

C ha-1) from PM soils (84 Mg C ha-1). PM soils had an average of 7.71 Mg N ha-1 in the

0-40cm depth, while AG soils contained 5.54 Mg N ha-1 at these depths; a 28% reduction.

PM soils had four times as much MBC in the soil surface as AG fields, 257 µg C g-1 soil compared to 64 µg C g-1 soil. PM soils had increased SOC levels in all particle size fractions. Clay-sized particles were observed as the dominant fraction of SOC in both soils.

In the short term component of the conversion study, a no-till approach to the land conversion was applied in an attempt to control the SOC disturbance caused by tillage. A replicated complete block (n = 3) experiment was established and plots in the prairie meadow (PM) were converted to wheat production (NT) through herbicide application in the summer of 2004.

Soil core samples were collected to a depth of 1m in May and June of 2008.

Management effects on SOC pools were assessed by analyzing total soil organic C

(SOC), total soil nitrogen (TSN), microbial biomass C (MBC) and a particle size fractionation of SOC in coarse sand (>250µm), fine sand (250-53µm), silt (53-2µm), and clay (<2µm) sized fractions. Total SOC, TSN, and all particle size fractions did not show any significant decreases, 3 growing seasons after the no-till conversion. MBC showed a significant decrease in the NT plots to a depth of 40cm. The decrease in microbial activity may be due to large losses of root biomass and the application of agronomic management. These results demonstrated that a no-till conversion of landscapes reduces

iv the impact on SOC pools, but more labile pools such as the MBC were heavily affected and may be predictive of future degradation in the system.

Together these studies provide further evidence that perennial plant communities store and cycle C, and maintain ecosystem processes at far greater levels than annual plant communities.

v DEDICATION

I would like to dedicate this thesis to

My wife and best friend Kat , and to

My parents Martha and Bill

vi ACKNOWLEDGEMENTS

In submitting this thesis and completing the requirements of the masters of science degree in the School of Environment and Natural Resources at the Ohio State

University, I would like to thank all those who have helped me through the work of the last 2 years. I thank my advisor, Prof. Rattan Lal, for giving me the opportunity to work and study at the Carbon Management and Sequestration Center (CMASC). Dr Lal has provided me with support, advice, an excellent work environment and he has taught me a tremendous amount about soil science, agriculture, and working for the environment. Dr Lal, I have tremendous gratitude and appreciation for all that you have done for me. I would like to thank the other members of my research committee, Dr.

Frank Calhoun and Dr. Martin Shipitalo, for their time, their advice and their support through the research process. I would also like to thank Dr Jerry Glover for the opportunity to do soil research at the Land Institute. Jerry has been an extremely helpful mentor.

I would like to thank all of my colleagues who have helped me with my work at

CMASC. I have been fortunate to spend the 2 years of my masters program with a great group of fellow students who have helped me to navigate graduate school and scientific research. Thank You Ji Young, Paula, Sindhu, and Umakant for all of your help. It has been a pleasure getting to know you all. A number of research scientists in Kottman Hall vii have also been extremely generous with advice and assistance during my work here. I would like to thank David Ussirri, Klaus Lorenz, and Nicola Lorenz for their helpful advice. I thank Sandy Jones for all of his advice and for his generosity with lab equipment. I thank Basant Rimal for all of the help that he has provided in the lab. I am grateful to Theresa Colson, Kate Elder, and the staff members of Kottman Hall 210 for everything that they have done for me in my time at Ohio State. I would also like to thank Emily Meyer at the Prairie Nature Center, for making my work there possible.

Finally, I am incredibly grateful to my wife Kat Deaner for the loving support and the wonderful home that she has given me. I could not have done this without you Kat.

I also thank the rest of my family for the support and friendship that they have given me during this process.

viii VITA

February 10, 1978 Born- Youngstown, Ohio

1996-2001…………………………...B.S. Plant Biology, Ohio University

2001-2002……………………………VISTA, Rural Action Sustainable Forestry Program, Rutland, Ohio

2002-2003……………………………Staff, Belize Agroforestry Research Center

2003-2007……………………………Principal, Habitats Landscaping

2007-present……………………….. Research and Teaching Assistant, Carbon Management and Sequestration Center, Ohio State University

FIELDS OF STUDY

Major Field: Environment and Natural Resources

Specialization: Soil Science

ix TABLE OF CONTENTS

ABSTRACT...... ii

DEDICATION...... vi

ACKNOWLEDGEMENTS...... vii

VITA ...... ix

TABLE OF CONTENTS ...... x

LIST OF TABLES………………………………………………………………………xvi

LIST OF FIGURES……………………………………………………………………xviii

1. INTRODUCTION...... 1

1.1 Soils and land use in the Global C Cycle...... 1

1.2 SOC and soil quality...... 5

1.3 Physical Fractionation of Organic Matter...... 6

1.4 Current Study ...... 8

References...... 9

2. CARBON SEQUESTRATION AND SOIL QUALITY IMPROVEMENT

POTENTIAL OF RESTORED TALLGRASS PRAIRIE IN OHIO ...... 12

2.1 Abstract...... 12

2.2 Introduction...... 13

2.3 Materials and Methods...... 18

2.3.1 Study Site ...... 18 x 2.3.2 Experimental Design and Soil Sampling ...... 19

2.3.3 Soil Analysis ...... 20

2.3.31 Bulk Density...... 20

2.3.3.2 Carbon and Nitrogen Analysis ...... 20

2.3.3.3 Aggregate Stability and Size Distribution...... 21

2.3.3.4 Soil Moisture Characteristics...... 22

2.3.3.5 Particulate Organic Matter Carbon ...... 23

2.3.3.6 Soil Particle Size Distribution ...... 23

2.3.4 Data Analysis ...... 25

2.4 Results and Discussion ...... 25

2.4.1 Soil Organic C ...... 25

2.4.2 Total Nitrogen...... 29

2.4.3 Carbon to Nitrogen Ratio ...... 33

2.4.4 Bulk Density...... 36

2.4.5 Water Stable Aggregation ...... 40

2.4.6 Aggregate Mean Weight Diameter...... 45

2.4.7 Available Water Capacity ...... 49

2.4.8 Total Porosity ...... 52

2.4.9 Coarse Particulate Organic Matter C ...... 58

2.4.10 Coarse Particulate Organic Matter Nitrogen ...... 62

2.4.11 Particle Size Distribution ...... 66

2.4.12 Soil Quality...... 68

2.5 Conclusions ...... 69 xi References...... 71

3. THE LONG-TERM EFFECTS OF THE CONVERSION FROM TALLGRASS

PRAIRIE TO WHEAT PRODUCTION ON SOIL ORGANIC CARBON IN NORTH

CENTRAL KANSAS...... 75

3.1 Abstract...... 75

3.2 Introduction...... 76

3.3 Materials and Methods...... 82

3.3.1 Study Site ...... 82

3.3.2 Soil Sampling ...... 84

3.3.3 Total C and N Concentrations ...... 84

3.3.4 Microbial Biomass C ...... 85

3.3.5 Particle Size Fractionation...... 86

3.3.6 Data Analysis...... 87

3.4 Results and Discussion ...... 88

3.4.1 Soil organic C ...... 88

3.4.2 Total soil N...... 94

3.4.2 Microbial Biomass C ...... 98

3.4.5 Particle Size Fractionation ...... 101

3.4.5.1 The Coarse Sand Fraction ...... 101

3.4.5.2 The Fine Sand Fraction ...... 105

3.4.5.3 The Silt Fraction ...... 109

3.4.5.4 The Clay Fraction...... 113

3.4.5.5 Relative Composition of SOC by the Particle Size Fractions ...... 117 xii 3.5 Conclusions ...... 119

References:...... 120

4. THE SHORT TERM EFFECTS OF THE CONVERSION FROM TALLGRASS

PRAIRIE TO WHEAT PRODUCTION ON SOIL ORGANIC CARBON IN NORTH

CENRAL KANSAS...... 124

4.1 Abstract...... 124

4.2 Introduction...... 125

4.3 Methods...... 128

4.3.1 Study Site ...... 128

4.3.2 Soil Sampling ...... 129

4.3.3 Total C and Total N ...... 131

4.3.4 Microbial Biomass C ...... 131

4.3.5 Particle Size Fractionation ...... 133

4.3.6 Data Analysis...... 134

4.4 Results...... 134

4.4.1 Soil Organic Carbon ...... 134

4.4.2 Total Soil Nitrogen ...... 138

4.4.3 Microbial Biomass C ...... 141

4.4.4Particle Size Fractionation ...... 145

4.4.4.1 The Coarse Sand Fraction...... 145

4.4.4.2 The Fine Sand Fraction ...... 147

4.4.4.3 The Silt Fraction ...... 150

4.4.4.4 The Clay Fraction...... 153 xiii 4.5 Conclusions ...... 156

References...... 156

COMPLETE REFERENCE LIST…………………………………………………...…159

APPENDIX A...... 166

xiv LIST OF TABLES

Table

1.1 Best Management Practices for C Sequestration in Agricultural Soils……….….3

2.1 Ohio Prairie Project Site Description……………………………………………19

2.2 Depth Distribution of SOC Concentration………………………………………27

2.3 Depth Distribution of Total Soil Nitrogen Concentration……………………….30

2.4 Depth Distribution of Carbon to Nitrogen Ratio………………………………...34

2.5 Depth Distribution of Soil Bulk Density………………………………………...37

2.6 Depth Distribution of Water Stable Aggregates…………………………………42

2.7 Depth Distribution of Aggregate Mean Weight Diameter………………………46

2.8 Depth Distribution of Available Water Capacity………………………………..50

2.9 Depth Distribution of Total Porosity……………………………………………54

2.10 Depth Distribution of Coarse Particulate Organic C……………………………59

2.11 Depth Distribution of Coarse Particulate Organic N……………………………63

2.12 Depth Distribution of Coarse Particulate Organic Matter C/N Ratio…………...65

2.13 Particle Size Distribution………………………………………………………..67

2.14 Critical Levels of Soil Physical Properties…………………………………… 69

3.1 Soil Taxonomic Descriptions of Research Sites………………………………...83

xv 3.2 Depth Distribution of SOC Pools…………………………………………….….92

3.3 Depth Distribution of TSN Pools…………………………………………….….96

3.4 Depth Distribution of MBC……………...……………………………………..100

3.5 Depth Distribution of SOC in the Coarse Sand Fraction……………………….102

3.6 Depth Distribution of SOC in the Fine Sand Fraction………………………….106

3.7 Depth Distribution of SOC in the Silt Fraction…………………………...……110

3.8 Depth Distribution of SOC in the Clay Fraction…………………………..…...114

4.1 Management of No-till Conversion Plots 2003-2007……………………….….130

4.2 Depth Distribution of SOC Pools in the Conversion Plots……………………..136

4.3 Depth Distribution of TSN Pools in the Conversion Plots……………………..139

4.4 Depth Distribution of MBC in the Conversion Plots…………………………...144

4.5 Depth Distribution of SOC in the Fine Sand Fraction in the Conversion Plots..148

4.6 Depth Distribution of SOC in the Silt Fraction in the Conversion Plots……….150

4.7 Depth Distribution of SOC in the Clay Fraction in the Conversion Plots…...…153

xvi LIST OF FIGURES

Figure

2.1 Depth Profile of SOC Concentration………………………………………….…28

2.2 SOC Rate of Change…………………………………………………………….29

2.3 Depth Profile of Total Soil Nitrogen Concentration…………………………….31

2.4 Simple Linear Regression of Total Soil N by SOC……………………………..32

2.5 Depth Profile of C/N Ratio……………………………………………………...35

2.6 Depth Profile of Soil Bulk Density…………………………………………...…38

2.7 Simple Linear Regression of Bulk Density by SOC………………………….…39

2.8 Rate of Change in Soil Bulk Density…………………………………………….40

2.9 Depth Profile of Water Stable Aggregates………………………………………43

2.10 Rate of Change in %WSA……………………………………………………….44

2.11 Depth Profile of Aggregate Mean Weight Diameter………………………….....47

2.12 Rate of Change in Aggregate Mean Weight Diameter……………………...…..48

2.13 Depth Profile of Available Water Capacity……………………………………..51

2.14 Rate of Change in Available Water Capacity……………………………….…..52

2.15 Depth Profile of Total Porosity………………………………………………….55 xvii 2.16 Simple Linear Regression of Total Porosity by Bulk Density………………..…56

2.17 Simple Linear Regression of Total Porosity by SOC……………………………56

2.18 Rate of Change in Total Porosity…………………………………………….….57

2.19 Depth Profile Coarse Particulate Organic Matter C…………………………..…60

2.20 Rate of Change in Total Porosity……………………………………………..…61

2.21 Depth Profile of Coarse Particulate Organic Matter N……………………..….64

3.1 Depth Profile of Soil Organic C Concentration……………………………..…89

3.2 Depth Profile of SOC in the Prairie Meadows of the Long-term Study Sites…90

3.3 Depth Profile of SOC in the Wheat Fields of the Long-term Study Sites……..91

3.4 Depth Profile of SOC Pools…………….………………………………..…….93

3.5 Depth Profile Total Soil N Concentration…………………………..……….....95

3.6 Depth Profile of TSN Pools………………………………………..…………..97

3.7 Depth Profile of Microbial Biomass C……………………………..………….99

3.8 Depth Profile of SOC in the Coarse Sand Fraction…………….…………….103

3.9 Depth Profile of TSN in the Coarse Sand Fraction……………………….…..104

3.10 Depth Profile of SOC in the Fine Sand Fraction…………………………...... 107

3.11 Depth Profile of TSN in the Fine Sand Fraction…………………………..….108

3.12 Depth Profile of SOC in the Silt Fraction………………..………………...... 111

3.13 Depth Profile of TSN in the Silt Fraction…………………………..……...…112

3.14 Depth Profile of SOC in the Clay Fraction………………………………...…115

3.15 Depth Profile of TSN in the Clay Fraction………………………………...…116

3.16 Relative Composition of Total SOC by SOC in Particle Size Fractions in the

Prairie Meadows…………………………………………………………...…117 xviii 3.17 Relative Composition of Total SOC by SOC in Particle Size Fractions in the

Wheat Fields……………………………………………………………….....118

4.1 Depth Profile of Soil Organic C Concentration in the Conversion Plots...…..135

4.2 Depth Profile of SOC Pools in the Conversion Plots……………………..….137

4.3 Depth Profile of TSN Concentration in the Conversion Plots…………...…...138

4.4 Depth Profile of TSN Pools in the Conversion Plots………………………....140

4.5 Depth Profile of Microbial Biomass C in the Conversion Plots……………...143

4.6 Depth Profile of SOC in the Coarse Sand Fraction in the Conversion Plots…145

4.7 Depth Profile of TSN in the Coarse Sand Fraction in the Conversion Plots…146

4.8 Depth Profile of SOC in the Fine Sand Fraction in the Conversion Plots……147

4.9 Depth Profile of TSN in the Fine Sand Fraction in the Conversion Plots…....149

4.10 Depth Profile of SOC in the Silt Fraction in the Conversion Plots……….....151

4.11 Depth Profile of TSN in the Silt Fraction……………………………….……152

4.12 Depth Profile of SOC in the Clay Fraction……………………………….….154

4.13 Depth Profile of TSN in the Clay Fraction………………………………..…155

xix CHAPTER 1

INTRODUCTION

1.1 Soils and land use in the Global C Cycle Atmospheric concentrations of radiatively active greenhouse gases (GHG’s) have risen dramatically in the last century (IPCC 2007). This has caused widespread concern

about long-term climate change. Carbon dioxide (CO2) is the largest component of greenhouse gas emissions, comprising 77% of total anthropogenic emissions in 2004

(IPCC 2007). Anthropogenic CO2 emissions result from a range of activities including: fossil fuel combustion, deforestation, biomass burning, cultivation of soils, drainage of wetlands, and other land use conversions (Lal 2008). These activities represent anthropogenic alteration of the fluxes, or transfer of C among pools, in the global C cycle.

The global C cycle is comprised of five major pools: oceanic (38,400 Pg), geologic or fossil (4, 130 Pg), atmospheric (760 Pg), biotic (560 Pg), and the pedologic or soil pool (2500 Pg) (Lal 2008). The soil pool contains more than twice the level of C in either the atmospheric or biotic pools. Estimates of the global soil pool suggest that there are 1550 Pg of organic C (SOC) and 950 Pg of inorganic C (SIC). These figures do not, however, account for the large quantities of SOC in permafrost soils that may be as high

1 as 495 Pg of SOC, in the top meter of those soils (Tarnocai et al. 2009). Soil has the potential to be a source or sink of C emissions, depending on management, as it is in active exchange with the atmospheric pool (Schimel 1995). Agricultural soils are of particular interest, because they are actively managed.

Land use changes from natural to managed ecosystems and the introduction and intensification of agriculture have accounted for large quantities of C to become oxidized

and emitted as CO2. Lal et al. (1999) estimated that the conversion of natural ecosystems to agricultural systems in the U.S., between 1750 to 1950, caused the depletion of 3 to 5

Pg of C from cropland soils, and the IPCC (2001) has estimated that, globally, most croplands have lost C on the order of 20- 40 Mg/ha. Researchers have found that grassland and forest soils generally lose somewhere between 20-50% of their original

SOC within the first 40 to 50 years of cultivation (Davidson and Ackerman 1993,

Houghton 1995). After this initial loss of SOC, changes in SOC content become largely a function of management practices in agricultural lands (Lal et al. 1999 and Follett

2001).

SOC stocks will be increased by practices that result in the amount of C entering the soil exceeding C lost to the atmosphere by oxidation (Follett 2001). In broad terms, strategies that can achieve this C sequestration through increasing SOC in agricultural lands include improved tillage practices and cropping systems, increased land cover, and efficient use of nutrient and water inputs (Follett 2001). It has been estimated that U.S. croplands have the potential to sequester C in the range of 5000 MMT by the year 2050; if best management practices (BMP’s) become widely applied (Table 1.1) (Lal et al.

2 1999). Land management that establishes and preserves perennial vegetation systems, such as forests, grasslands, and wetlands also sequesters C in the SOM.

BMP’s for Optimizing Soil C Sequestration in Agricultural Soils

-Incorporation of Crop Residues - Conservation Tillage - Fallow - Perennial Vegetation (CRP Land) - Crop Rotations - Cover Crops - Precision use of Inorganic N Fertilizers - Manure/Biosolids

Table 1.1: Best Mangement Practices for C Sequestration in Agricultural Soils from Follett (2001) and Lal et al. (1999).

Currently, 1.6 Pg yr-1, of the total 8.6Pg yr-1, of anthropogenic C emissions are the result of land use changes, primarily the conversion of native ecosystems to agricultural systems (IPCC 2001; Lal 2008). Returning some of these agricultural areas to biomass production systems based on perennial plant communities is one strategy that can be effective at offsetting some portion of anthropogenic emissions by sequestering SOC

(Post and Kwon 2000). A meta-analysis of studies documenting changes in SOC following land use conversions found that converting agricultural soils to perennial 3 grasslands and to secondary forests were the most C negative land use options (Guo and

Gifford 2002). If managed for biomass production, perennial plant based systems may offer multiple benefits, as perennial systems have demonstrated the ability to maintain long term ecosystem services at a far higher level than annual crop based systems (Glover et al. 2009).

As humanity enters the 21st century with more than 6.5 billion people, the world’s soils will be managed to meet increasing demands including: increased food production, increased biomas production, the restoration of degraded lands, C sequestration, increased resource use efficiency in agriculture, and preservation of biodiversity (Lal

2007a). Agro-ecosystems centered around perennial plants may offer a key component to meeting humanities goals while regenerating soils. There is currently a high-level of scientific interest in perennial production systems as society looks to the landscape to produce huge quantities of plant-based bio-fuel, while maintaining ecosystem services

(Robertson et al. 2008). Reviews comparing several bio-fuel production systems have indicated that systems composed of perennial plants, such as prairie grasses, store greater levels of SOC (Anderson et al. 2009) and offer the most C negative land-use conversion options (Fargione et al. 2008), of all the systems examined.

Field trials in the Midwestern US have confirmed these results. At the Kellogg

Biological Station in Michigan, field plots under perennial alfalfa and young secondary forest have been accumulating SOC at more than twice the rate as plots under annual crops (Robertson and Grandy 2007). In Minnesota, low-input high diversity prairie plantings have demonstrated the ability to produce large quantities of biomass on degraded soils, and to sequester more SOC than the equivalent fossil C emitted during 4 their biofuel production process (Tilman et al. 2006). Annually harvested prairie meadows in Kansas have maintained higher biomass yields and similar nitrogen yields as conventionally farmed wheat fields on the same soils while receiving only 8% of the energy inputs, over the last 80 years (Glover et al. 2009).

1.2 SOC and soil quality. A central component in balancing productivity with ecological sustainability in agroecosystems is the management of soil quality. Soil quality is the capacity of a soil to sustain biological productivity, maintain environmental quality and promote plant and animal health (Doran and Parkin 1994). The soil quality concept is a scientific paradigm dealing with the associations of soil management practices, observable soil characteristics, soil processes, and soil’s level of ecosystem function (Lewandowski et al.

1999). Soil quality is typically assessed by measuring a number of soil parameters critical to ecosystem functioning, and comparing them among soils under differing management regimes (Doran and Parkin 1994). This includes measuring soil physical, chemical, and biological properties and using robust statistical analysis to demonstrate differences.

Soil organic C is a key indicator of overall quality. SOC is central to many of the ecosystem processes facilitated by the soil (Lal 2007b). SOC has a large influence over many specific soil properties that are critical for soil quality including: soil aggregation, soil water availability, cation exhange capacity and nutrient availability, microbial biomass C, and pH buffereing (Weil and Magdoff 2004). Integrated crop management

(Glover et al. 2000), organic management (Reaganold et al. 1993), reduced tillage

5 (Wander and Bollero 1999) and the retention of crop residues (Karlen et al. 1994) are all management strategies that have proven effective at improving soil quality in agricultural soils. Water stable aggregation (McQuaid and Olson 1998) and particulate organic matter (POM), both heavily influenced by SOC, have been identified as particularly sensitive indicators of soil quality.

1.3 Physical Fractionation of Organic Matter In order to understand the effects of land management on SOC, researchers have had to develop concepts and methodologies for describing the dynamics of C in the soil.

Conceptually, SOC is thought to be contained in three kinetically defined pools: a labile pool with short turnover times, an intermediate pool, and a more stable, and humified passive pool with long turnover times (Parton et al. 1987). While these pools have offered a valuable conceptual tool for the study of SOC dynamics, they have proven elusive in the field and laboratory. Physical fractionation methods seek to isolate and analyze physically distinct fractions of SOC in the laboratory. Physical fractionation studies have become increasingly widespread due to an increased awareness that SOC turnover is a biological process that is regulated by the physical structure of the soil

(Christensen 2001). SOC turnover is regulated not only by the chemical nature of the substrate itself, but through the nature of its association with the mineral components of the soil (Sollins 1996, Christensen 2001). These studies suggest 3 levels of physical association and regulation between SOC and the mineral components of the soil: 1) organomineral complexes between SOC and primary soil mineral particles, 2) the

6 secondary aggregate structure of the soil, and 3) in the structurally intact whole soil

(Christensen 2001).

The study of SOC associated with primary particle size separates in the soil offers valuable insight to management induced changes in SOC and SOC turnover. Particle size fractionation begins with the complete dispersion of the soil, either through chemical

(Na-hexametaphosphate (HMP)) or physical (sonication, shaking) means, or some combination of the 2 (Christensen 1992). Sand size particles are separated from the resulting slurry by wet or dry sieving approaches, and silt and clay size particles are separated through sedimentation or centrifugation procedures (Christensen 1992).

Additionally, density-based separation of light fraction organic C (LFOC) in the sand sized separates is commonly included in the fractionation process. Studies that use physical fractionation methods to separate the sand sized particulate organic matter fraction (POM) are also widespread.

SOC in primary particle size separates has demonstrated distinct chemical and ecological differences among fractions. A review of dozens of studies that used primarly particle size fractionation indicated that biological activity and C/N ratios increased with increasing particle size, while turnover time and 14C age increased with decreasing particle size (Lutzow et al. 2007). SOC within the sand fraction tends to be described as the active pool, while SOC in the silt and sand size fractions is thought of as being in the intermediate and passive pools, though this is considered a rough approximation (Lutzow

2007). The association between SOC and reactive minerals, such as clay, silt and oxides, is a primary stabilization pathway for SOC (Sollins et al. 1996). The total mineral surface area of the silt and clay size particles, as determined by particle size 7 distribution, has demonstrated accuracy as a measure or a soil’s total protective capacity

(Hassink 1997) or saturation level (Six et al. 2002) for SOC. The principal is that soils with greater total silt and clay composition, and surface area, have increased ability to stabilize SOC.

While silt and clay associated SOC has shown slow turnover times, the sand sized

POM has proven responsive to short term changes and has become a valuable tool in evaluating soil management impacts. POM-C has been correlated with numerous soil properties and functions, including: POM was found to be a high proportion of the SOM oxidized by the initial cultivation of grasslands (Cambardella and Elliot 1992); POM has shown strong correlations with organic N mineralization (Wilson et al. 2001); POM is used to show subtle differences in the effects of agronomic management on SOM (Sa et al. 2001 and Marriott and Wander 2006); and POM-C has strong correlations with soil quality (Wander and Bollero 1999).

1.4 Current Study This study has focused on SOC changes in land management systems based around tallgrass prairie in the Midwestern U.S. The first component of the study examined changes in SOC and soil physical properties in a restored prairie chronosequence in northwest Ohio. The second component of the study examined the long-term and short-term effects of the conversion of tallgrass prairie to wheat production on SOC, in north central Kansas. The study is centered around the following objectives

1) To evaluate the changes in SOC under restored tallgrass prairie.

8 2) To evaluate the effect of planting tallgrass prairies on soil quality, in

previously cultivated soils.

3) To evaluate the long-term effects of the conversion of tallgrass prairie to

wheat production on SOC fractions in north central Kansas.

4) To evaluate the short-term effects of a no-till conversion of tallgrass prairie to

wheat production on SOC fractions in north central Kansas.

Chapter 2 of this thesis focuses on the study done in the restored prairie chronosequence in Ohio. Research objectives 1 and 2 are addressed in Chapter 2. Chapter 3 of this thesis presents the study on the effects of the conversion of tallgrasss prairie to wheat production on SOC fractions in Kansas, and addresses research objective 3. Chapter 4 of the thesis presents the study of the no-till conversion experiment in Kansas, and will address research objective 4.

References Batjes, N.H. 1996. Total carbon and nitrogen in the soils of the world. European Journal of Soil Science 47: 151-163. Christensen, B.T. 1992. The physical fractionation of soil and organic matter into primary particle size and density separates. Advances in Soil Science 20: 1-90. Christensen, B.T. 2001. Physical fractionation of soil and structural and functional complexity in organic matter turnover. European Journal of Soil Science 52: 345- 353. Davidson, E.A., and Ackerman, I.L. 1993. Changes in soil carbon inventories following the cultivation of previously untilled soils. Biogeochemistry 20: 161-193. Doran, J.W. and Parkin, T.B. 1994. Defining and Assessing Soil Quality. In Doran, J.W., Coleman, D.C., Bezdicek, D.F., and Stewart, B.A. (Eds). Defining Soil Quality for a Sustainable Environment. Soil Science Society of America, Inc., Madison, WI.

9 Fargione, J., Hill, J., Tilman, D., Polasky, S., and Hawthorne, P. 2006. Land clearing and the biofuel carbon debt. Science 319: 1235-1237. Follett, R.F. 2001. Soil Management Concepts and Carbon Sequestration in Cropland Soils. Soil and Tillage Research. 61: 77-92. Glover, J.D., Culman, S.W., DuPont, S.T., Broussard, W., Young, L., Mangan, M.E., Mai, J.G., Crews, T.E., DeHaan, L.R., Buckley, D.H., Ferris, H., Turner, R.E., Reynolds, H.L., Wyse, D.L., 2009. Harvested perennial grasslands provide ecological benchmarks for agricultural sustainability (submitted to Agriculture, Ecosystem, and Environment). Glover, J.D., Reganold, J.P., and Andrews, P.K. 2000. Systematic method for rating soil quality of conventional, organic and integrated apple orchards in Washington state. Agriculture, Ecosystems, and Environment 80: 29-45. Grandy, A.S. and Robertosn, G.P. 2007. Land-use intensity effects on soil organic carbon accumulation rates and mechanisms. Ecosystem 10: 58-73. Guo, L.B. and Gifford, R.M. 2002. Soil carbon stocks and land use change: a meta- analysis. Global Change Biology 8: 345-360. Hassink, J. 1997. The capacity of soils to preserve organic C and N by their association with clay and silt particles. Plant and Soil 191: 77-87. Houghton, R.A. 1995. Changes in storage of terrestrial carbon since 1850. in R. Lal, J. Kimble, E. Levine, and B.A. Stewart (eds) “Soils and Global Change”. Boca Raton, FL: CRC Publishers. IPCC. 2007. Climate Change 2007: Synthesis Report. Fourth assessment report of the Intergovernmental Panel on Climate Change. IPCC. 2001. Climate Change 2001: The Scientific Basis. Intergovernmental Panel on Climate Change. Karlen, D.L., Wohllenhaupt, N.C., Erbach, D.C., Berry,, E.C., Swan, J.B., Eash, N.S. and Jordahl, J.L. 1994. Crop residue effects on soil quality following 10-years of no- till corn. Soil and Tillage Research 32: 313-327. Lal, R. 2008. Carbon Sequestration. Philosphical Transactions of the Royal Society, Biological Sciences, 363: 815-830. Lal, R. 2007a. Soils and sustainable agriculture: A review. Agronomy for Sustainable Development 27: 1-8. Lal, R. 2007b. Soil Science and the Carbon Civilization. Soil Science Society of America Journal 71: 1425-1437. Lal, R., Kimble, J.M., Follett, R.F., and Cole, C.V. 1999. The Potential of U.S. Cropland to Sequester Carbon and Mitigate the Greenhouse Effect. Boca Raton, FL. CRC Press. Lewandowski, A., Zumwinkle, M., and Fish, A. 1999. Assessing the soil system: A review of soil quality literature. Minnesota Department of Agriculture, Energy and Sustainable Agriculture Program, St. Paul, MN. Marriott, E.E. and Wander, M. M. 2006b. Total and Labile Soil Organic Matter in Organic and Conventional Farming Systems. Soil Science Society of America Journal 70:950-959. McQuaid, B.F., and Olson, G.L. 1998. Impact of carbon sequestration on functional I

10 indicators of soil quality. In Lal, R., Kimble, J.M., Follett, R.F., and Stewart, B.A. (Eds) Soil Processes and the Carbon Cycle. CRC Press, Boca Raton, FL. Parton, W.J., Brookes, P.C., Coleman, K., and Jenkinson, D.S. 1987. Dynamics of C, N, S, and P in grassland soils: a model. Biogeochemistry 5: 109-131. Post, W.M. and Kwon, K.C. 2000. Soil carbon sequestration and land-use change: process and potential. Reganold, J.P., Palmer, A.S., Lockhart, J.C., and Macgregor, A.N. 1993. Soil quality and financial performance of biodynamic and conventional farms in New Zealand. Science 260: 344-349. Sa, J.C. de M., Cerri, C. Dick, W.A., Lal, R. Filho, S.P.V. , Piccolo, M.C. and Feigl. B.E. 2001. Organic Matter Dynamics and Carbon Sequestration Rates for a Tillage Chronosequence in a Brazilian Oxisol. Soil Science Society of America Journal 65:1486-1499. Schimel, D.S. 1995. Terrestrial ecosystems and the carbon cycle. Global Change Biology 1: 77-91. Six, J., Conant, R.T., Paul, E.A., and Paustian, K. 2002. Stabilization mechanisms of soil organic matter: implications for C-saturation of soils. Plant and Soil 241: 155-176. Sollins, P., Homann, P., and Caldwell, B.A. 1996. Stabilisation and destabilization of soil organic matter: mechanisms and controls. Geoderma 74: 65-105. Tarnocai, C., Canadell, J.G., Schuur, E.A.G., Kuhry, P., Mazhitova, G., and Zimov, S. 2009. Soil organic carbon pools in the northern circumpolar permafrost region. Global Biogeochemical Cycles 23: GB2023. Tilman, D., Hill, J. and Lehman, C. 2006. Carbon-negative biofuels from low-input high diversity grassland biomass. Science 314: 1598-1600. Wander, M.M., and Bollero, G.A. 1999. Soil Quality Assessment of Tillage Impacts in Illinois. Soil Science Society of America Journal 63: 961-971. Weil, R.R., and Magdoff, F. 2004. Significance of soil organic matter to soil quality and health. In Magdoff, F. and Weill, R.R. (Eds) Soil Organic Matter in Sustainable Agriculture. CRC Press, Boca Raton, FL. Wilson, T.C., Paul, E. A. and Harwood, R.R. 2001. Biologically active soil organic matter fractions in sustainable cropping systems. Applied Soil Ecology 16: 63

11 CHAPTER 2

CARBON SEQUESTRATION AND SOIL QUALITY IMPROVEMENT

POTENTIAL OF RESTORED TALLGRASS PRAIRIE IN OHIO

2.1 Abstract

This study analyzed soils in a restored tallgrass prairie landscape with the primary objectives of: 1) To assess the changes in soil organic C (SOC) in restored prairie landscapes and to compare them with the SOC in conventional agricultural systems on the same soil, and 2) To quantify the impact of the restored prairie plantings on soil quality parameters and compare it with soils under conventional agricultural land uses.

The primary study site was the Prairie Nature Center at the Marion Campus of The Ohio

State University. The Prairie Nature Center (PNC) is a 4.5 ha site that began the creation/restoration of tallgrass prairie plantings in 1977. Soil samples were taken from

31year (P77), 13 year (P95), and 8 year old (P00) restored prairie plantings, as well as from lawn (LA) and an annually cultivated corn (Zea mays)/soybean (Glycine max) field

(AG) on similar soils.

Soil samples were taken from 0-10, 10-20, 20-30, and 30-40cm depth, from 4 sites in each treatment during the summer of 2008. These soils demonstrated significant increases in SOC concentration, particulate organic matter C (POM-C), water stable 12 aggregation (%WSA), aggregate mean weight diameter (MWD), total porosity (ft), and available water capacity (AWC), and significant decreases in soil bulk density (ρb) associated with time under tallgrass prairie. The accrual of SOC, in the previously cultivated soil, was greatest in the surface layer (0-10cm) of the P77 soil (3.45%), compared with the AG soil (2.14%). Soil aggregate properties showed significant increases in the P00 (95.48%WSA) from the AG treatment (71.11%WSA). This indicates that macro-aggregate properties may be restored relatively quickly (<10yrs) under restored tallgrass prairie. The majority of significant changes observed in this study occurred in the soil surface (0-10cm) layer, which may be due to the high proportion of prairie root biomass in soil surface layers.

2.2 Introduction The management of vegetation and soil has a significant role to play in strategies directed at offsetting anthropogenic impact on the global C cycle. Of the approximately

8.5 Pg C yr –1 currently being emitted by human activity into the earth’s atmosphere, no less than 1.5 Pg C yr –1 are emitted from the management and degradation of land, primarily through the conversion of native ecosystems to agriculture (IPCC 2007). In temperate soils, researchers have documented that grassland and forest soils typically lose between 20-50% of their soil organic C (SOC) within 40 to 50 years of the onset of cultivation (Houghton 1995). Alternatively, land management practices can also be adopted to sequester C, or take it out of the atmospheric C pool and store it in soil and vegetation C pools (Lal 2008). Soils represent a significant component of terrestrial C

13 management, as the world’s soils contain a larger reservoir of organic C than both the atmosphere and the vegetation of the planet combined (Lal 2008).

The SOC stocks in agricultural soils can also be increased by best management practices (BMPs) that result in the amount of C entering the soil exceeding C lost to the atmosphere by oxidation (Follett 2001). It has been estimated that U.S. croplands have the potential to sequester C in the range of 5000 MMT by the year 2050; if BMPs were to be widely applied (Lal et al. 1999). One of the most effective management decisions for

C sequestration, is to allow degraded agricultural soils to revert to natural vegetation, or to replant perennial vegetation systems (Post and Kwon 2000). Numerous researchers have demonstrated high rates of C sequestration on lands enrolled in the USDA’s

Conservation Reserve Program (CRP); which provides incentives to farmers who take land out of production and establish perennial vegetation cover (Follett 2001; Lal et al.,

1999).

This strategy of converting land from annual crop production to perennial vegetation seems particularly appropriate for degraded agricultural soils. These are the soils which have a reduced ability to provide ecosystem services and crop production

(Lal 2004). Soil degradation often occurs as agricultural intensification and tillage break down the soil’s physical structure, oxidize SOC and increase the soil’s susceptibility to water and wind erosion. A review of the data on soil degradation in the US suggests that there are on the order of 400,000 ha of degraded cropland in a seven state region surrounding Ohio (Lal et al. 2004).

Rather than allowing soil degradation, land managers must strive to restore soil quality; or the capacity of a soil to sustain biological productivity, maintain 14 environmental quality, and promote plant and animal health (Doran and Parkin 1994).

Robust assessment of soil quality is carried out by measuring a suite of soil parameters, critical to ecosystem functioning, and comparing them among soils managed differently

(Doran and Parkin 1994). The SOC concentration is a principle indicator of soil quality, as many ecosystem processes and services are directly mediated by SOC levels (Lal

2007). These ecosystem services, provided by healthy soils, are essential for the continuing prosperity of the agricultural communities around them.

Tallgrass prairie is a perennial vegetation system that may be particularly well suited to restoring SOC and soil quality in marginal agricultural soils through use as a long-term fallow, a perennial biomass production system (Tilman et al 2006), or in conservation buffer systems (Blanco-Canqui et al. 2004). Tallgrass prairie is a highly productive, native North American plant community dominated by warm season, perennial C4 grasses; that also includes forbs, legumes, and cool season C3 grasses

(Knapp and Seastedt 1998). Tallgrass prairie ecosystems are characterized by having a large allocation and storage of C in belowground biomass and SOC pools (Buyanovsky et al 1987, Rice et al 1998). This tremendous allocation of C into belowground pools results in prairies having higher SOC levels than perennial vegetation systems dominated by trees in similar climatic conditions (Seastedt and Knapp 1993). Created tallgrass prairies are those which have been planted, either as restored prairie in areas where native prairies grew historically, or as created prairies in landscapes without a history of tallgrass prairies. The creation of tallgrass prairies, and other native ecosystems, on degraded landscapes offers numerous ecosystem and societal services including: preservation of biodiversity, wildlife habitat, soil conservation, and C sequestration. 15 Studies of restored prairies in the Midwestern US have shown that the planting of prairies can sequester C as SOC. In a study over 30 restored prairies in CRP lands in

Minnesota, the results demonstrated that prairies were sequestering linear rates of total

SOC, light fraction C (LF-C) and non-hydrolyzable SOC over time, and suggested that prairie restoration builds SOC on a decadal time-scale (McLauchlan et al. 2006). A long-term study of a restored prairie chronosequence in Illinois indicated linear increases in SOC pools and changes in soil physical properties that correlate with the longevity of the prairie plantings (Matamala et al. 2008, Jastrow 1996). This study suggested that restored prairies could regain SOC levels of around 50% of their historical levels after

100yrs and that prairie restoration is a viable strategy for soil C sequestration (Matamala et al. 2008). Conversely, studies on prairie restoration chronosequences in Wisconsin suggest low rates of SOC accumulation in restored prairies (Brye and Kucharik 2003,

Kucharik et al. 2006). No studies currently exist on the rate of SOC accumulation in restored prairies in Ohio, the eastern limit of the historical tallgrass prairie range.

Tallgrass prairie may have increasing importance in the American agricultural landscape as a biomass crop. The 2007 US Farm Bill has mandated that by 2022 the US will produce 60 billion liters of cellulosic ethanol annually. The enormous pressure that this level of production will place on the landscape requires a strong research focus on ecosystem services in biomass production systems (Robertson et al 2008). Prairie grasses, such as switchgrass (Panicum virgatum), are capable of producing large quantities of net cellulosic energy per unit area (Schmer et al 2008). Other researchers have reported that biologically diverse prairie plantings are able to produce more biomass and SOC than grass monocultures (Tilman et al. 2006, Fornara and Tilman 2008). In 16 Kansas, prairie meadows where biomass has been harvested annually for a century support significantly higher levels of ecosystem services than adjacent annual croplands

(Glover et al. 2009). Other studies on the net C balances of biofuel production systems have suggested that converting agricultural land to perennial grassland offers a strong positive C sequestration benefit (Fargione et al. 2008), while systems already in CRP perennial vegetation will have a negative net C balance for approximately 50 years if converted to corn (Zea mays) ethanol production (Pineiro et al. 2009).

Tallgrass prairie plantings have also been shown to have a strong effect on restoring the physical structure of cultivated soils. Jastrow (1987) observed that soil macro-aggregate formation and stabilization occurred rapidly after the planting of tallgrass prairie in the Illinois chronosequence. Macroaggreagate formation at this site was later calculated to reach 95% of historic prairie equilibrium just 10 years after restoration (Jastrow 1996). These data suggest that macroaggregate formation under tallgrass prairie plantings is facilitated by the extensive prairie root systems. Several studies in restored and native tallgrass prairies have indicated correlations between the adundance of mycorrhizal fungi and macro-aggregate formation (Wilson et al. 2009).

This study evaluated a tallgrass prairie restoration chronosequence in northwestern Ohio with the objectives: 1) to assess the SOC levels in a restored prairie chronosequence and compare them with adjacent agricultural and lawn areas, and 2) to quantify and compare the impact of prairie restoration on soil quality parameters. The soil quality parameters analyzed in the study are primarily soil physical properties. They include SOC, Total N, water stable aggregation (%WSA), aggregate mean weight

diameter (MWD), total porosity (ft), available water capacity (AWC), coarse particulate 17 organic matter C (CPOM-C), and coarse particulate organic matter nitrogen (CPOM-N).

The hypotheses being tested are: 1) SOC levels will increase with time under tallgrass prairie, and 2) Soil physical quality parameters will improve with time under tallgrass prairie.

2.3 Materials and Methods

2.3.1 Study Site This study was conducted at the Prairie Nature Center (PNC) on the Marion campus of The Ohio State University. The PNC was begun with the planting of a 0.40 ha tallgrass prairie in 1977 in a field that had previously been under long term cultivation.

The PNC now has a chronosequence of created prairies ranging in age from 1 yr to 31yrs.

Areas adjacent to the PNC are still used for corn (Zea mays) and soybean (Glycine max) cultivation. Created prairies have been grown from both seeds and container grown plants, and the prairie areas have been burned every other year since 1995. Fire is a key component of prairie ecology and the practice of burning the prairie plantings gives the native prairie plant species competitive advantage over exotics and keeps tree species from colonizing the plantings.

The prairie plantings at the site contain high levels of plant diversity, consisting primarily of native perennial grasses and forbs. Dominant grasses in the areas sampled included the following species: Andropogon gerardii, Panicum virgatum, Sorgahastrum nutans, and Schizachyrium scoparium. The following forb species were also found in abundance in the sampling areas: Asclepias tuberosa, Asclepias syriaca, Baptisia leucantha, Desmodium canadense, Echinacea purpurea, Helianthus grosseratus, 18 Heliopsis helianthoides, Liatris spicata, Mondarda fistulosa, Ratibida pinnata, Silphium integrifolium, Silphium perfoliatum, Silphium trifoliatum, Silphium terebinthinaceum, and Solidago canadensis. Mean annual temperature at the site is 10.4 °C and mean annual precipitation is 864mm.

Ohio Prairie Project Site Description

Research Site Soil Type Treatments

The Prairie Nature Center Pewamo 31 Year Prairie (P77)

N 40°34’52” 0-2% Slope 13 Year Prairie (P95)

W 85°5’21” 8 Year Prairie (P00)

Corn/Soy Agriculture (AG)

Lawn (LA)

Table 2.1: Ohio Prairie Project Site Description

2.3.2 Experimental Design and Soil Sampling Soil samples were collected in August 2008. Four sample replicates were taken in each of the treatment areas: a 31 year old prairie (P77), a 13 year old prairie (P95), an

8 year old prairie (P00), an adjacent corn/soybean field (AG), and an adjacent lawn area

(LA). All samples were collected from the fertile, bottomland Pewamo series (Fine, mixed, active, mesic Typic Argiaquoll) on glacial till parent material. Soil samples were taken at 0-10, 10-20, 20-30, and 30-40cm depths. Intact soil cores, taken with a hammer driven probe, and bulk soil samples taken at each depth.

19 In the lab, clods from the bulk soil were broken along weak planes at field moisture content and sieved through an 8mm sieve. Clods between 8-4.75mm were separated from the bulk soil for use in aggregate analysis and all samples were air-dried at room temperature. Bulk soil samples were passed through a 2mm sieve before analysis.

2.3.3 Soil Analysis 2.3.31 Bulk Density

Intact core samples were used to determine soil bulk density (ρb) following the method of Blake and Hartge (1986). The cores used were approximately 7.5cm in diameter by 7.5cm in height. Cores were weighed at field moisture to calculate the wet

bulk density (ρ′b) by dividing the total mass of the soil by the total volume of the soil.

25g sub-samples samples of field moist bulk soil were then dried in an oven at 105°C

overnight to determine the gravimetric moisture content (w). Dry bulk density (ρb) of the soil was then calculated using the formula:

ρb = ρ′b / (1 + w)

2.3.3.2 Carbon and Nitrogen Analysis Total C and Total N were analyzed by combustion analysis. Dry samples were passed through a 2mm sieve, ground with a mortar and pestle, and roller ground to achieve a fine and homogenous mixture. Samples were analyzed for total soil C and N concentration by dry combustion at 900°C (Nelson and Sommers 1996) using a CN

Elemental Analyzer (Vario Max, Elementar Analysensysteme, Hanau, Germany).

20 2.3.3.3 Aggregate Stability and Size Distribution A wet sieving procedure was used to determine the stability and size distribution of soil aggregates following a method first outlined by Yoder (1936). Aggregates in the size range of 4.75 to 8mm were obtained by gently breaking soil clods along weak planes at field moisture content. These samples were then air dried for analysis.

Fifty grams of dry soil aggregates were placed on a set of five nested sieves of

4.75, 2, 1, 0.5, and 0.25 mm diameter. The aggregates were then pre-wetted by capillarity on the upper (4.75mm) sieve for 30min. The nest of sieves were then oscillated vertically with an amplitude of 3cm at a rate of 2 oscillations per second for 30 min. The soil aggregate fractions retained on each sieve were then washed into pre- weighed 250ml beakers and oven dried at ~60° C for 48h and weighed, to determine the mass of each size fraction. The mass of aggregates in the <.25mm fraction was then obtained by taking the difference between the initial sample mass and the sum or the masses recorded for the 5 sieve fractions.

Water stable aggregation (%WSA) and mean weight diameter (MWD) were then calculated by the methods of Kemper and Rosenau (1986) and used as indices of soil structure.

The %WSA, or % of total aggregation (>0.25mm) was calculated using the equation:

%WSA >.25mm = 100*(M1 + M2 + M3 + M4 + M5) / M0

where M1-5 are equal to the mass of soil aggregates recovered on each of the sieves and

M0 is equal to the initial mass of soil used for the wet sieving procedure.

21 Aggregate mean weight diameter (MWD) was calculated with the equation:

n MWD = Σ j=1 xjmj

Where n = the number of aggregate size ranges (mm), mj = the mass of the aggregates of the given size range as a fraction of the total dry mass of the sample analyzed, and xj = the mean diameter of any particular size range of aggregates separated by sieving.

2.3.3.4 Soil Moisture Characteristics The intact soil core samples were stored in plastic bags at field moisture content, after weighing for bulk density. Cores were trimmed at both ends and cheesecloth was used to secure the bottom of the cores. To prepare for moisture retention analysis, the cores were pre-wetted by capillarity for 24h. Soil moisture characteristics were determined at 0, -3, and –6 kPa with a tension table and at –10, -30, -100, -300, and

–1500 kPa with pressure plate extractors following methods of Klute (1986). The gravimetric soil moisture content (w) was converted to volumetric moisture content (θ) by multiplying with the dry soil bulk density.

Total porosity (ft) was calculated to be the same as the volumetric moisture content (θ) at saturation (0 kPa). Available water content (AWC) was calculated by subtracting the θ at the permanent wilting point (PWP: -1500 kPa) from the θ at the field capacity (FC: -300 kPa). This θ was then multiplied by the depth, in cm, to achieve a

AWC value in cm of available water.

22 2.3.3.5 Particulate Organic Matter Carbon Bulk soil samples were analyzed for C and N concentrations in the coarse particulate organic matter fraction (CPOM) using the method described by Sollins et al.

(1999). Briefly, 10 g of sieved (2mm), air-dried soil were mixed with 30 ml of

Hexametaphosphate solution (5 g L-1) in a 250ml bottle and shaken overnight (18 h) in a floor shaker. The soil suspension was then washed through at 53 µm sieve with deionized water. The material retained on the sieve was washed with deionized water into a pre-weighed beaker and oven dried at 45°C. The dried mass of the sand fraction was the then weighed and finely ground for dry combustion analysis of total C and N concentrations. CPOM-C and N concentrations ( g C or N kg-1 soil) were then calculated using the following equation:

CPOM-C = Cs x Ws x 10

where Cs is the %C of the sand fraction and Ws is the dry mass of the sand fraction (g fraction/g soil).

2.3.3.6 Soil Particle Size Distribution Soil particle size distribution was measured using the hydrometer method. The hydrometer method is based on Stoke’s Law, or the fact that particles of different sizes have different settling velocities in liquid. Briefly, 51g of air-dry, sieved (2mm) soil was weighed and transferred to a metal mixing cup. 50ml of 0.2M Na-hexametaphosphate were added to the cup along with 75ml of deionized water. The sample was then mixed with a stirring rod and left to soak for 30min. The sample cup was then transferred to the multi-mix machine and stirred for 15min. After stirring the soil was transferred from the

23 mixing cup to a graduated settling cylinder. Deionized water was then added to the cylinder until the water level reached 1000ml.

Once filled, the cylinder top was covered by hand, and the cylinder was fully inverted 5 times. After the fifth inversion, the cylinder was placed on the lab table, a stop

watch was initialized and the time was recorded as time0 . A hydrometer was then placed floating in the cylinder. The first hydrometer reading was taken at 40s on the stop watch

and recorded as H1, the temperature was taken and recorded as T1. After the readings the hydrometer was removed and the sample was allowed to continue settling. A second

hydrometer reading, H2, and a second temperature reading, T2, were taken at 3 hours from the time0.

The temperature and hydrometer readings were then used to calculate the particle size distribution, or percent sand, silt and clay. The first set of readings is used to determine the percent sand, while the second set of readings is used to determine percent clay, and the silt is determined by subtracting the sum of the sand and clay from 100%.

The following equations were used for the calculations.

Sand = 100 – [H1 + 0.2(T1 - 68°) – 2.0] 2.0

Clay = [H2 – 0.2(T2 - 68°) – 2.0] 2.0

Silt = 100 – (%sand + %clay)

Where H1 is the first hydrometer reading and T1 is the first temperature reading, and H2 is the second hydrometer reading and T2 is the second hydrometer reading. These equations calibrate the hydrometer readings according to the temperature at the time of the reading.

24 2.3.4 Data Analysis Results were analyzed using a 1-way analysis of variance (ANOVA) testing the effect of treatment at each depth using JMP version 7 statistical software (SAS Institute,

Cary, N.C.). Mean values from the ANOVA were then tested for statistical differences using Tukey’s HSD (Honest Significant Difference) test, at α = 0.05.

2.4 Results and Discussion 2.4.1 Soil Organic C The SOC concentration showed large differences related to treatment primarily in the surface (0-10cm) layer. The 31-year prairie (P77) had the highest concentration of SOC

(3.45%) in the surface layer. The two younger prairies, P95 and P00, also had high levels of SOC at 2.92% and 2.86%, respectively. The agricultural treatment (AG) had 2.14%

SOC and the lawn treatment had 1.80% SOC. The P77 treatment was significantly different than both the lawn and the agriculture treatments (Table 2.2), and all three of the prairie treatments had statistically higher SOC levels than the lawn treatment. The P95 and P00 treatments were not significantly different than the AG treatment, and the AG treatment showed no significant difference over the LA treatment.

Below the surface layer, the SOC concentrations do not follow the predictable treatment driven pattern of the surface soils. Beginning in the 10-20cm layer, the 8-year prairie (P00) has the highest level of SOC concentration at 2.37%. This P00 treatment continues to have the highest levels of SOC through the sampling profile to 40cm depth.

This is likely due to the P00 treatment’s position in the bottom of the landscape at the

Prairie Nature Center. The differences in SOC concentration that do exist below the 0- 25 10cm layer, though not statistically significant, are all likely due to background differences relating to subtle changes that occur in slope and landscape position among the treatment areas. All treatments showed SOC levels between 1.5% and 2%, above

30cm, indicating that the poorly drained, bottomland soils at the site are relatively high in organic matter content.

The lawn (LA) treatment shows continually lower levels of SOC than the other treatments, despite being sampled in a similar landscape position. This observed SOC deficit is likely due to some historical disturbance of the soil, probably associated with the construction of the campus and the adjacent parking lot. It is quite possible that topsoil was lost from the lawn area, making comparisons about SOC concentration in that treatment difficult.

Using the agricultural treatment as a baseline, SOC concentration has increased at a rate of 0.04% SOC yr-1 (0.4g C kg-1 soil yr-1), during the 31 years that the P77 treatment has been in place. The rate of increase is greater, at 0.09% SOC yr-1 (0.9g C kg-1 soil yr-

1), if the 8-year timeframe between the P00 and baseline is considered. SOC appears to have a linear rate of increase (Fig 2.2) and the rate appears to be continuing in the 31 year old treatment (P77).

26

Soil Organic Carbon Concentration (%)

Depth (cm) P77 P95 P00 LA AG

0-10 3.45(.23) A 2.92(.23) AB 2.86(.23) AB 1.80(.23) C 2.14(.23) BC

10-20 1.85(.19) - 2.05(.19) - 2.37(.19) - 1.54(.19) - 2.02(.19) -

20-30 1.73(.18) - 2.00(.18) - 2.21(.18) - 1.45(.18) - 1.86(.18) -

30-40 1.79(.27) - 1.67(.27) - 1.98(.27) - 1.17(.27) - 1.18(.27) –

Table 2.2: Depth Distribution of SOC Concentration (%). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Values in parentheses are standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

27 Soil Organic C (%) 0 0.5 1 1.5 2 2.5 3 3.5 4

A AB 10 AB C BC

20 P77 P95 P00 LA AG

Depth (cm) 30

40

Figure 2.1: Depth Profile of SOC Concentration (%). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Error bars represent standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

28 4.50

4.00

3.50

3.00

2.50 0-10cm Linear (0-10cm) 2.00 SOC (%) 1.50

1.00

0.50

0.00 0 5 10 15 20 25 30 35 Time (yrs)

Figure 2.2 SOC Rate of Change SOC in the soil surface (0-10cm) of plots of different age in the chronosequence.

2.4.2 Total Nitrogen Total soil N (TSN) values showed a low level of significant differences among treatments. The prairie treatments had greater TSN values than the LA and AG treatments in the surface soil layer (0-10cm). The 13 year-old prairie (P95) had the highest value at 0.26%. The P77 and P00 treatments both had a TSN value of 0.24%, the

AG treatment had a value of 0.21% and the LA treatment was 0.18% TSN. The only statistically significant difference in TSN among treatments in the surface (0-10cm) layer

29 was the difference between P95 (0.26%) and LA (0.18%). Below 10cm depth, there were no significant differences in TSN among the treatments and there was little pattern to the distribution of TSN (Fig 2.3). Regression analysis of %N by %C shows that there is a strong, positive linear relationship (r2 = 0.81) between N concentration and SOC concentration, suggesting that much of the N in these soils is associated with SOC.

Total Soil Nitrogen (%)

Depth (cm) P77 P95 P00 LA AG

0-10 .24 (.02) AB .26 (.02) A .24 (.02) AB .18 (.02) B .21 (.02) AB

10-20 .15 (.02) - .18 (.02) - .20 (.02) - .17 (.02) - .19 (.02) -

20-30 .14 (.02) - .20 (.02) - .18 (.02) - .16 (.02) - .19 (.02) -

30-40 .15 (.02) - .16 (.02) - .18 (.02) - .13 (.02) - .14 (.02) -

Table 2.3: Depth Distribution of Total Soil Nitrogen Concentration (%). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Values in parentheses are standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

30 Total Soil N (%) 0 0.05 0.1 0.15 0.2 0.25 0.3

AB A 10 AB B AB

20 P77 P95 P00 LA

Depth (cm) 30 AG

40

Figure 2.3: Depth Profile of Total Soil Nitrogen Concentration (%). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Error bars represent standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

31 Linear Fit: % N = 0.057 + 0.063*% C, r2 = 0.81

Figure 2.4: Simple Linear Regression of Total Soil N (%) by SOC (%), run in JMP 7.

32 2.4.3 Carbon to Nitrogen Ratio The data on C/N ratio show widening C/N ratios with the accumulation of SOC in the surface soil (0-10cm) of the prairie treatments. The P77 treatment surface layer showed the highest recorded C/N ratio at 14.21, which was significantly different than any of the other treatments. This high C/N ratio is likely due to the accumulation of plant residues on the soil surface after 31 years of prairie growth. Another factor may be the burning that occurs regularly in the prairies. Burning may lead to the deposition of charcoal and black C residues with wide C/N ratios on the soil surface. The P00 surface soil had the next highest C/N value at 12.15, also significantly different from all other treatments. The three prairie treatments have C/N values that are consistently higher than the LA and AG treatments throughout the 40cm sampling depth.

The LA and AG treatments had C/N ratios of 10.16 and 10.34, respectively, in their surface soils. These values are significantly lower than the prairie treatments. The

LA and AG treatments had C/N ratios below 10 in the 10-40cm depths. These values were lower than any of the values measured in the prairie treatments. These low C/N ratios may reflect two differences between the LA and AG systems from the prairie areas.

There is likely far less belowground biomass in both the LA and AG systems compared with the prairies. The LA and AG systems are both receiving N fertilizer inputs on an annual basis, while the prairies are not.

33

C/N Ratio

Depth (cm) P77 P95 P00 LA AG

0-10 14.21(.30) A 11.34(.30) BC 12.15(.30) B 10.16(.30) C 10.34(.30) C

10-20 12.29(.47) A 11.01(.47) AB 11.76(.47) A 8.96(.47) B 10.50(.47) AB

20-30 12.08(.36) A 10.18(.36) B 11.93(.36) A 9.33(.36) B 9.88(.36) B

30-40 12.06(.63) A 10.26(.63) AB 11.04(.63) AB 8.59(.63) B 8.34(.63) B

Table 2.4: Depth Distribution of Carbon to Nitrogen Ratio. Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Values in parentheses are standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

34 C/N Ratio 0 5 10 15 20

A BC 10 B C C A AB P77 20 A B P95 AB P00 A B LA 30 Depth (cm) A B AG B A AB 40 AB B B

Figure 2.5: Depth Profile of C/N Ratio. Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Error bars represent standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

35 2.4.4 Bulk Density

Soil bulk density (ρb) showed little variation among treatments. The only ρb value that was statistically significant from others was the P77 surface sample, which had a value of 1.11g cm-3. This is likely due to the large changes in soil structure and accumulation of SOC that has occurred with 31 years of prairie growth. The P95 and

P00 samples also showed trends toward slightly lower bulk density in surface soils with bulk density values of 1.32 and 1.29g cm-3, compared with the LA (1.36g cm-3) and AG

(1.46g cm-3). The samples taken below the soil surface showed fairly high bulk density values in all treatments ranging from 1.46 to 1.57g cm-3 in samples taken in the 10-40cm depths. This may be due to reduced porosity and increased clay concentration, occurring with depth.

Similar patterns in ρb were observed in a prairie restoration chronosequence in

Illinois, where ρb values showed their greatest response to treatment in the uppermost

15cm layer (Matamala et al. 2008).

Regression analysis of ρb by SOC concentration (Fig. 2.7) shows a moderate negative linear correlation between the two properties (r2 = 0.58). This suggests that changes in these 2 properties are related at this site. The general trend that was observed was that of decreasing bulk density with increasing SOC. There was a fairly linear rate of decrease in bulk density over time (Fig 2.8) in the chronosequence. Bulk density showed a rate of decrease of .01g soil cm-3 yr-1 over the 31-year period since the planting of P77. This rate was slightly elevated during the first 8 years after planting with bulk

-3 -1 density decreasing .02 g soil cm yr between the AG (t0) and the P00 (8yr) treatment. It

36 appears that the trend in bulk density is that it is still decreasing (Fig 2.8) in these soils after 31 years

Bulk Density (g cm-3)

Depth (cm) P77 P95 P00 LA AG

0-10 1.11(.04) B 1.32(.04) A 1.29(.04) A 1.36(.04) A 1.46(.04) A

10-20 1.47(.03) - 1.49(.03) - 1.49(.03) - 1.57(.03) - 1.57(.03) -

20-30 1.48(.03) - 1.45(.03) - 1.46(.03) - 1.55(.03) - 1.54(.03) -

30-40 1.49(.05) - 1.50(.05) - 1.48(.05) - 1.52(.05) - 1.54(.05) -

Table 2.5: Depth Distribution of Soil Bulk Density (g cm-3). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Values in parentheses are standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

37 Dry Bulk Density (g/cm3) 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8

B A 10 A A A

20 P77 P95 P00 LA 30 AG Depth (cm)

40

Figure 2.6: Depth Profile of Soil Bulk Density (g cm-3). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Error bars represent standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

38 Linear Fit: BD = 1.75 - 0.14*% C, r2 = 0.58

Figure 2.7: Simple Linear Regression of Bulk Density (g cm-1 ) by SOC (%).

39 1.60

1.40

1.20

1.00

0-10cm 0.80 Linear (0-10cm)

0.60

0.40 Bulk Density (g / cm3) 0.20

0.00 0 5 10 15 20 25 30 35 Time (yrs)

Figure 2.8 Rate of Change in Soil Bulk Density Bulk density in the soil surface (0-10cm) of plots of different age in the chronosequence.

2.4.5 Water Stable Aggregation All samples in this study had relatively high values for %WSA (> 71.11%). The 3 prairie treatments all had %WSA values above 95% in the top 20cm, with P77 having above 95% WSA throughout the sampling profile. While these values did not show a high degree of statistical significance (Table 2.6), they represent a very high level of soil secondary structure. The levels of aggregation in the prairie treatments were also higher than those observed in the LA and AG treatments throughout the sampling profile. The lowest %WSA value measured in the study was in the surface (0-10cm) of the AG soil. 40 This is likely due to the long history of surface cultivation in that soil, as cultivation of grassland soils is known to greatly reduce macro-aggregation (Gupta and Germida 1988,

Cambardella and Elliot 1993). The high levels of WSA measured in all treatments are likely the result of the relatively high SOC levels of the Pewamo soil series.

The %WSA values measured in the prairie treatments in this study are in line with a series of studies that assessed aggregation in a restored prairie chronosequence in Illinois

(Jastrow, 1987, Jastrow 1996). These studies indicated that planting tallgrass prairie in cultivated soils restored WSA to an equilibrium level, compared with undisturbed prairie remnants, after approximately 10.5 years (Jastrow 1996). The author hypothesized that the formation of macro-aggregates in the prairie system facilitated the further accrual of

SOC and the formation of micro-aggregates (Jastrow 1996). Other studies on restored prairies (Miller and Jastrow 2000) and in remnant prairies (Wilson et al. 2009) have shown a strong correlation between macro-aggregates and the presence of mycorrhizal fungi in prairies. It may be that the extensive root system and abundant mycrorrhizal fungi in tallgrass prairies are responsible for the high levels of macroaggregation found in prairie restoration sites.

The rate of change in %WSA (Fig 2.10) has demonstrated a pattern of high levels of initial increase in aggregation in the first 8 years under prairie with the increases leveling off after that and %WSA remaining fairly constant at approximately 95%. This level of aggregation (95%) may represent an equilibrium level in these soils. There was a very

marked increase between the AG (t0) treatment at 71%WSA and the P00 (8 year) treatment at 95.48% WSA, which amounts to an increase of 3% annually.

41

Water Stable Aggregates (%)

Depth (cm) P77 P95 P00 LA AG

0-10 96.99(4.24)A 96.39(4.24)A 95.48(4.24)A 86.25(4.24)AB 71.11(4.24)B

10-20 96.74(3.90)A 95.23(3.90)A 95.53(3.90)A 76.58(3.90)B 81.24(3.90)AB

20-30 95.74(3.61)A 94.56(3.61)A 87.03(3.61)AB 78.20(3.61)B 81.91(3.90)AB

30-40 95.87(3.65)- 90.39(3.65)- 93.72(3.65)- 84.38(3.65)- 82.09(3.65)-

Table 2.6: Depth Distribution of Water Stable Aggregates (%). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Values in parentheses are standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

42 Water Stable Aggregates (%) 0 20 40 60 80 100 120

A A 10 A AB B A A 20 A P77 B AB P95 P00 A A LA 30 AB AG

Depth (cm) B AB

40

Figure 2.9: Depth Profile of Water Stable Aggregates (%). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Error bars represent standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

43 120

100

80

60 0-10cm %WSA (%) 40

20

0 0 5 10 15 20 25 30 35 Time (yrs)

Figure 2.10: Rate of Change in %WSA %WSA in the soil surface (0-10cm) of plots of different age in the chronosequence.

44 2.4.6 Aggregate Mean Weight Diameter Aggregate Mean Weight Diameter followed a pattern similar to the one observed in

%WSA analysis with high MWD values in all of the prairie treatments. The P77 and P95 treatments had MWD values that were statistically greater than the AG treatment in the soil surface (0-10cm). In the 10-20cm and 20-30cm depths, all 3 of the prairie treatments had significantly higher MWD values that the LA and AG treatments. These results reflect the same processes of macro-aggregation, described above with the %WSA results, that occur in the presence of the prairie’s extensive root system.

The changes in MWD observed in the chronosequence (Fig 2.12) parallel those

observed in %WSA. The MWD values increase dramatically from the AG (t0) at

1.78mm to the P95 (13 year) treatment at 5.00mm. This is a rate of increase of .25mm yr-

1. After the P95 observation, the MWD values appear to level off and remain constant, with an MWD observation of 5.11mm in the P77 (31yr) treatment. A MWD value of

5mm may represent an equilibrium level in these soils.

45

Mean Weight Diameter (mm)

Depth (cm) P77 P95 P00 LA AG

0-10 5.11(.48)A 5.00(.48)A 3.88(.48)AB 3.54(.48)AB 1.78(.48)B

10-20 4.91(.37)A 4.45(.37)A 4.48(.37)A 2.26(.37)B 2.15(.37)B

20-30 4.82(.35)A 3.97(.35)A 3.82(.35)A 2.03(.35)B 1.92(.35)B

30-40 4.51(.40)A 3.52(.40)AB 3.90(.40)AB 1.27(.40)C 2.34(.40)BC

Table 2.7: Depth Distribution of Aggregate Mean Weight Diameter (mm). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Values in parentheses are standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

46 Mean Weight Diameter (mm) 0 1 2 3 4 5 6

A A 10 AB AB B A A P77 20 A B P95 B P00 A A LA 30 Depth (cm) A AG B B A AB 40 C AB BC

Figure 2.11: Depth Profile of Aggregate Mean Weight Diameter (mm). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Error bars represent standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

47 6.00

5.00

4.00

3.00 0-10cm

MWD (mm) 2.00

1.00

0.00 0 5 10 15 20 25 30 35 Time (yrs)

Figure 2.12: Rate of Change in Aggregate Mean Weight Diameter MWD in the soil surface (0-10cm) of plots of different age in the chronosequence.

48 2.4.7 Available Water Capacity The plant available water capacity, or AWC, did not show a high degree of change in this study. The P77 surface (0-10cm) sample had the highest recorded AWC value, at

1.9cm. This value was significantly higher than the AG treatment, but not statistically different than the other treatments. AWC values were highest in the surface layer.

There, the 3 prairie treatments and the LA treatment all showed improvement when compared to the AG treatment. The P77 and P95 treatments both showed gradual increases in AWC with decreasing depth. The changes in AWC in the surface layer are likely due to the accumulation of large levels of SOC observed in those soils. Increasing

SOC is known to increase both the FC and the AWC (Hudson 1994).

AWC showed continuing increases over time (Fig2.14), in the chronosequence measurements. The trend in increases in AWC seems to be described with the linear trend, though there is a wide variation in the AWC of the 31yr (P77) treatment. It also appears that the AWC is continuing to increase in these soils. The average rate of

increase in AWC from the AG (t0) treatment to the P77 (31yr) treatment is approximately

.025cm yr-1 of available water.

49

Available Water Capacity (cm)

Depth (cm) P77 P95 P00 LA AG

0-10 1.9(.20) A 1.4(.20) AB 1.6(.20) AB 1.5(.20) AB 1.1(.20) B

10-20 1.2(.10)- 1.0(.10) - 1.0(.10) - 1.2(.10) - .90(.20) -

20-30 1.0(.10) - 0.9(.10) - 1.0(.10) - 1.3(.10) - 1.1(.10) -

30-40 0.80(.20) - 0.7(.20) - 1.1(.20) - 1.1(.20) - .90(.20) –

Table 2.8: Depth Distribution of Available Water Capacity (cm). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Values in parentheses are standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

50 Available Water Capacity (cm) 0 0.5 1 1.5 2 2.5

A AB 10 AB AB B

20 P77 P95 P00 LA 30 AG Depth (cm)

40

Figure 2.13: Depth Profile of Available Water Capacity (cm). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Error bars represent standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

51 0.30

0.25

0.20

0-10cm 0.15 Linear (0-10cm) AWC (cm) 0.10

0.05

0.00 0 5 10 15 20 25 30 35 Time (yrs)

Figure 2.14: Rate of Change in Available Water Capacity AWC in the soil surface (0-10cm) of plots of different age in the chronosequence.

2.4.8 Total Porosity

Total porosity (ft) levels recorded in this study were strongly affected by management in the soil surface (0-10cm). The P77 treatment had the highest recorded value at .56 cm3 cm-3, which was statistically different from all other treatments. The AG treatment had the lowest recorded value at .37 cm3 cm-3. Both the P77 and P00 treatments had values that were statistically different from the LA and AG treatments. Porosity values did not show strong treatment effects below the surface layer. 52 Total porosity showed a strong, negative linear correlation with soil bulk density (Fig

2.16, r2= 0.75) and a positive linear correlation with %SOC (Fig 2.17, r2=0.63), suggesting that changes in total porosity are related to long term changes in these two key soil parameters. The strong correlation between the porosity data and the bulk density data suggests confidence in the data, as these two related parameters were measured using different methodologies in this experiment. Regression analysis of porosity with

%WSA showed a low level of correlation (r2= 0.15), given that both parameters are an index of the soil’s secondary structure.

Total porosity showed continuous increases over time (Fig 2.18), in the chronosequence. Porosity increased steadily from the AG treatment to the P77 treatment, though the rate was somewhat gradual with average increases of .006cm3 cm-3 yr-1 over

31 years. The trend in the data suggests that porosity is continuing to increase after 31 years (Fig 2.18).

53

Total porosity (cm3 cm-3)

Depth (cm) P77 P95 P00 LA AG

0-10 .56(.02) A .46(.02) BC .47(.02) B .39(.02) CD .37(.02) D

10-20 .40(.02) AB .39(.02) AB .41(.02) A .33(.02) B .37(.02) AB

20-30 .39(.01) AB .41(.01) A .42(.01) A .33(.01) B .40(.01) A

30-40 .37(.02) - .39(.02) - .43(.02) - .34(.02) - .39(.01) –

Table 2.9: Depth Distribution of Total Porosity (cm3 cm-3). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Values in parentheses are standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

54 Total Porosity (cm3 cm-3) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

A BC 10 B CD D

AB AB 20 A P77 B AB P95 P00 AB LA A 30 A AG Depth (cm) B A

40

Figure 2.15: Depth Profile of Total Porosity (cm3 cm-3). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Error bars represent standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

55 Linear Fit: Total Porosity = 0.98 - 0.40*BD, r2 = 0.75

Figure 2.16: Simple linear regression fitting Total Porosity (cm3 cm-3) by Bulk Density (g cm-3).

Linear Fit: Total Porosity = 0.27 + 0.07*% C, r2 = 0.63

Figure 2.17: Simple linear regression fitting Total Porosity (cm3 cm-3) by SOC (%).

56 0.70

0.60

0.50

0.40 0-10cm Linear (0-10cm) 0.30

0.20

Total Porosity (cm3/cm3) 0.10

0.00 0 5 10 15 20 25 30 35 Time (yrs)

Figure 2.18: Rate of Change in Total Porosity Total porosity in the soil surface (0-10cm) of plots of different age in the chronosequence.

57 2.4.9 Coarse Particulate Organic Matter C The observed levels of CPOM-C demonstrated a strong treatment effect. The P77 treatment showed a high level of CPOM-C (14.09g C kg-1 soil) in the soil surface. This value was significantly different from all other treatments. The P77 and P00 treatments were significantly higher in CPOM-C in the 20cm layer than the other 3 treatments. All

3 of the prairie treatments showed significantly higher levels of CPOM, at 30cm depth, than the LA and AG treatments. The AG treatment consistently had the lowest levels of

CPOM observed in the study.

These CPOM-C results are consistent with previous studies of POM-C in both grassland and cultivated soils, where cultivated soils have shown a significant reduction in POM-C, as POM-C is believed to be highly affected by tillage (Cambardella and Elliot

1992). The relative contribution of CPOM-C to the total SOC was also greatly reduced in the AG treatment compared with the long-term prairie P77. CPOM-C in the P77 treatment (14.09g C kg-1 soil) represents 41% of total SOC (34.5g C kg-1 soil) in that treatment, while the CPOM-C in the AG treatment (3.29g C kg-1 soil) comprises 15% of total AG SOC (21.4g kg-1 soil). These results are also consistent with the Cambardella and Elliott (1992) study, where POM-C comprised 39% of total SOC in the native grassland treatment and 18% in a bare fallow cultivated treatment.

POM-C showed a fairly high rate of increase in samples from the chronosequence

(Fig??). POM-C also appears to be continuing to increase after 31 years, though there is wide variation in the levels of POM-C detected in the 31yr samples (Fig 2.20). POM-C increased from 3.29 g C kg-1 soil in the AG treatment to 14.09g C kg-1 soil in the P77 treatment, which represents a rate of 0.35g C kg-1 soil yr-1.

58

Coarse Particulate Organic Matter-C (g C kg-1 soil)

Depth (cm) P77 P95 P00 LA AG

0-10 14.09(1.19)A 5.95(1.19)B 6.21(1.19)B 4.11(1.19)B 3.29(1.19)B.

10-20 4.00(.20)A 2.27(.20)B 3.25(.20)A 2.26(.20)B 2.08(.20)B

20-30 2.71(.17)A 2.44(.17)A 2.66(.17)A 1.59(.17)B 1.49(.17)B

30-40 1.90(.19)AB 1.62(.19)BC 2.50(.19)A 1.14(.19)BC 1.05(.19)C

Table 2.10: Depth Distribution of Coarse Particulate Organic C (g C kg-1 soil). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Values in parentheses are standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

59 Coarse Particulate Organic Matter-C (g C/kg soil) 0 5 10 15 20

A B 10 B B B A B P77 20 A B P95 B P00 A A LA 30 A Depth (cm) B AG B AB BC 40 A BC C

Figure 2.19: Depth Profile Coarse Particulate Organic Matter C (g C kg-1 soil). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Error bars represent standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

60 25.00

20.00

15.00 POM-C Rate of Change

Linear (POM-C Rate of Change) 10.00

POM-C (g C/ kg soil) 5.00

0.00 0 10 20 30 40 Time (yrs)

Figure 2.20: Rate of Change in Total Porosity Total porosity in the soil surface (0-10cm) of plots of different age in the chronosequence.

61 2.4.10 Coarse Particulate Organic Matter Nitrogen C-POM N did not follow the same patterns in abundances as the CPOM-C. CPOM-N showed statistically significant results among treatments in the upper 20cm of the soil

(Table 2.11). In the top 10cm, P77 had significantly more POM-N (0.71g N kg-1 soil) than any of the other treatments. This result seems to reflect the quantity of POM in the

P77 treatment, as the C/N ratio of the CPOM in this soil (19.9) was significantly higher than any of the other treatments. The POM C/N ratio in the P77 treatment is significantly higher than any of the other treatments through 30cm depth (Table 2.11), demonstrating that after 31 years the prairie has POM residues that are very high in C. The P00 treatment had the highest levels of POM-N in the 10-20cm (0.22g N kg-1 soil) and in the

20-30cm depth (0.20g N kg-1 soil). The comparatively low POM C/N ratios in the AG treatment may reflect the input of N fertilizer.

62

Coarse Particulate Organic Matter-N (g N kg-1 soil)

Depth (cm) P77 P95 P00 LA AG

0-10 0.71(.07)A 0.40(.07)B 0.43(.07)AB 0.26(.07)B 0.25(.07)B

10-20 0.21(.01)AB 0.17(.01)B 0.22(.01)A 0.16(.01)B 0.17(.01)B

20-30 0.15(.01)BC 0.18(.01)AB 0.20(.01)A 0.12(.01)C 0.12(.01)C

30-40 0.13(.01)AB 0.13(.01)AB 0.18(.01)A 0.09(.01)B 0.09(.01)B

Table 2.11: Depth Distribution of Coarse Particulate Organic N (g N kg-1 soil). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Values in parentheses are standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

63 Coarse Particulate Organic Matter-N (g N/kg soil) 0 0.2 0.4 0.6 0.8 1

A B 10 AB B B

AB B 20 A P77 B B P95 P00 BC LA

Depth (cm) AB 30 A AG C C

AB AB 40 A B B

Figure 2.21: Depth Profile of Coarse Particulate Organic Matter N (g N kg-1 soil). Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Error bars represent standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

64

Coarse Particulate Organic Matter- C/N Ratio

Depth (cm) P77 P95 P00 LA AG

0-10 19.9(.52)A 14.85(.52)BC 14.4(.52)BC 15.85(.52)B 12.95(.52)C

10-20 19.2(.47)A 13.4(.47)BC 14.23(.47)BC 14.45(.47)B 12.23(.47)C

20-30 18.0(.28)A 13.35(.28)B 13.55(.28)B 13.53(.28)B 12.68(.28)B

30-40 14.75(.63)A 12.88(.63)AB 14.13(.63)AB 12.93(.63)AB 11.53(.63)B

Table 2.12: Depth Distribution of Coarse Particulate Organic Matter C/N Ratio. Values are means from a 1-way ANOVA testing the effect of treatment at each depth. Values in parentheses are standard errors. Capital letters indicate t-groupings, or groups that are significantly different, in Tukey’s Honest Significant Difference (HSD) test (∝ = 0.05).

65 2.4.11 Particle Size Distribution Based on the particle size distribution analysis (Table 2.13), the textural classes of the surface layer (0-10cm) are the following: P77 and P00 are loams, and P95, AG, and LA are clay loams. There was, however, a discrepancy between the amount of sand indicated by the hydrometer method and the amount of sand retained in the CPOM analysis. It may be that the CPOM method achieved a higher degree of dispersion. In the CPOM analysis, the soils were shaken over night in a HMP solution. In the hydrometer method, a briefer dispersion process was applied to the soils (section 2.3.6). This may have resulted in incomplete dispersion. It may be that some of the sand measured using the hydrometer method was actually microaggregates that were not fully dispersed.

66

Particle Size Distribution Sample %Sand %Clay %Silt CPOMSand

P77 10 44 11 45 30 P77 20 40 18 42 26 P77 30 38 20 42 23 P77 40 38 22 40 22

P95 10 32 25 43 23 P95 20 31 26 43 22 P95 30 32 27 41 21 P95 40 34 28 38 18

P00 10 38 17 45 23 P00 20 36 19 45 22 P00 30 35 21 44 21 P00 40 36 23 41 21

AG 10 32 23 45 18 AG 20 29 28 43 18 AG 30 28 29 43 16 AG 40 30 30 40 15

LA 10 32 24 44 20 LA 20 31 27 42 20 LA 30 31 23 46 19 LA 40 27 27 46 15

Table 2.13 Particle Size Distribution. Includes the %Sand, %Clay, and %Silt from the hydrometer method and the %Sand retained in the CPOM analysis.

67 2.4.12 Soil Quality Soil quality was evaluated for treatments of 3 different times since the planting of

tallgrass prairie in the study(Table 2.14): AG (t0), P00 (8yrs), and P77 (31yrs). A complex statistical index of soil quality has not been defined for the study, but instead observed values of soil physical parameters were evaluated against critical levels. In a method for assessing the sustainable management of soil and water resources, Lal (1994) defined critical limits as the level of a soil parameter beyond which crop production declines. Critical limits also define the severity of degradation of the soil resource (Lal

1994). Observed values for soil properties are assigned a value for their relation to the critical levels. Greater values indicate a higher level of soil degradation.

According to the Lal (1994) critical value rating system, none of the treatments measured fall into the category of extreme degradation. The AG treatment has a majority of 3’s assigned to soil physical parameters, indicating a moderate degree of soil limitation on the AG soil. The P00 treatment showed a slight degree of limitation, while the P77 treatment did not demonstrate any limitation. This analysis appears to support the hypothesis that the restored prairie plantings are able to restore soil physical quality.

68

Critical Levels of Soil Physical Properties Treatment %SOC B.D. %WSA MWD

AG 2.14 (3) 1.46 (3) 71.11 (2) 1.78 (3)

P00 2.86 (3) 1.29 (2) 95.48 (1) 3.88 (1)

P77 3.45 (2) 1.11 (1) 96.99 (1) 5.11 (1)

Table 2.14 Critical levels of soil physical properties Assigns a rank to observed soil parameter levels. Increasing rank values indicate greater soil degradation.

2.5 Conclusions Tallgrass prairie plantings showed a strong improvement, with age, in all of the soil parameters measured in this study. Prairie plantings increased SOC concentration, restored the macro-aggregation levels to very high levels (>95% WSA), and showed marked decreases in bulk density, large increases in total porosity, and four-fold increase in CPOM-C, when compared to the adjacent annual agriculture field. Macro-aggregation levels showed strong improvement in the P00 (8 year prairie treatment), while the other parameters measured showed their greatest increases in the oldest P77 prairie treatment.

Continuing increases were observed in the rates of change of SOC, AWC, total porosity, and POM-C, and continuing decreases in bulk density, through the 31-year treatment in the chronosequence. %WSA and MWD showed rapid rates of increase through the 8- year treatment and then appeared to reach an equilibrium state. A comparison of three

69 treatments with critical levels for soil physical parameters demonstrated a marked increase in overall soil physical quality with increased time under tallgrass prairie.

The majority of large changes in soil physical parameters observed in this study were observed in the surface 0-10cm layer. This may be due to the high levels of root biomass found in the surface of prairie soils. Review of the literature on tallgrass prairie root biomass suggests that a very large proportion of prairie roots occur in the surface, with studies observing: 44% of root biomass in 0-10cm (Rice et al. 1998), 80% in

0-25cm (Kucera and Dahlman 1968) and 80% in 0-15 (Matamala et al. 2008). The observation of dramatic changes in the soil physical structure in the soil surface may also have been driven by the deposition or large quantities of residue and the increases in biological activity associated with the tallgrass prairie.

The high level of water stable aggregation observed in the 8 year old (P00) prairie treatment demonstrates that tallgrass prairie plantings can begin restoring soil physical quality on a relatively short timescale. This observation is consistent with previous studies looking at macro-aggregates in restored tallgrass prairie (Jastrow 1996). That study indicated that macro-aggregation was restored to pre-cultivation levels in approximately 10.5 years (Jastrow 1996). The study then hypothesized that the formation of macroaggregates in those soils then worked to facilitate the further accrual of SOC

(Jastrow 1996). That theory is supported by the data generated in this study, where the 8 year prairie had already achieved 95%WSA and SOC continued to accumulate in the prairie soils, reaching 3.45% concentration in the P77 treatment.

The fourfold increase in CPOM-C from the AG treatment (3.29g C kg-1 soil) to the P77 treatment (14.09g C kg-1 soil) also seems quite significant. POM-C is a measure 70 of the accumulation of plant-derived residues in the soil and is a basic substrate for the soil food web. Previous studies of soil quality have found that POM-C showed the highest statistical correlation, of any parameter measured, with overall soil quality and suggested that it may be an early indicator of changes in soil quality (Wander and Bollero

1999). This study has shown that CPOM-C greatly increases with time under tallgrass prairie. The wide C/N ratio in the CPOM of the P77 treatment (19.9) suggests that this fraction can become increasingly deficient in N over time, under tallgrass prairie.

Overall, this study demonstrates that planting tallgrass prairie on agricultural soils can be an effective strategy for building SOC and improving soil quality. These results add to the case that perennial grasslands, such as prairies, may offer one of the most elegant solutions to the problem of increasing biomass production levels in the American agricultural landscape. Prairie plantings provide an effective choice for land managers seeking to improve soil quality and the concurrent ecosystem services on agricultural soils.

References Blake, G.R., Hartge, K.H., 1986. Bulk Density. In: Klute, A. (Ed.), Methods of Soil Analysis Part 1- Physical and Mineralogical Methods. Soil Science Society of America, Inc., Madison, WI, USA, pp. 364-367. Blanco-Canqui, H., Gantzer, C.J., Anderson, S.H., Alberts, E.E., and Thompson, A.L. 2004. Grass barrier and vegetative filter strip effectiveness in reducing runoff, sediment, nitrogen and phosphorous loss. Soil Science Society of America Journal 68: 1670-1678. Brye, K.R. and Kucharik, C.J. 2003. Carbon an nitrogen sequestration in two prairie topochronosequences on contrasting soils in southern Wisconsin. American Midland naturalist 149: 90-103. Buyanovsky, G.A., Kucera, C.L., and Wagner, G.H. 1987. Comparative Analysis of Carbon Dynamics in Native and Cultivated Ecosystems. Ecology 68: 2023-2031. Cambardella, C. A. and Elliott, E. T. 1992. Particulate soil organic matter changes

71 across a grassland cultivation sequence. Soil Science Society of America Journal 56: 777-783. Cambardella, C. A. and Elliott, E. T. 1993. Carbon and nitrogen distribution in aggregates from cultivated and native grassland soils. Soil Science Society of America Journal 57: 1071-1076. Doran, J.W. and Parkin, T.B. 1994. Defining and Assessing Soil Quality. In Doran, J.W., Coleman, D.C., Bezdicek, D.F., and Stewart, B.A. (Eds). Defining Soil Quality for a Sustainable Environment. Soil Science Society of America, Inc., Madison, WI. Fargione, J.J., Hill, J., Tilman, D., Polasky, S., and Hawthorne, P. 2008. Land clearing and the biofuel carbon debt. Science. 319: 1235-1238. Follett, R.F. 2001. Soil management concepts and carbon sequestration in cropland soils. Soil and Tillage Research 61: 77-92. Fornara, D.A. and Tilman, D. 2008. Plant functional composition influences rates of soil carbon and nitrogen accumulation. Journal of Ecology 96: 314-322. Glover, J.D. et al. 2009. Harvested perennial grasslands provide ecological benchmarks for agricultural sustinablility. Agriculture, Ecosystems, and Environment, Submitted. Gupta, V. V. S. R. and Germida, J. J. 1988. Distribution of microbial biomass and its activity in different soil aggregate size classes as affected by cultivation. Soil Biology and Biochemistry 20: 777-786. Houghton, R.A. 1995. Changes in storage of terrestrial carbon since 1850. in R. Lal, J. Kimble, E. Levine, and B.A. Stewart (Eds.) Soils and Global Change. CRC Publishers Boca Raton, FL. Hudson, B. D. 1994. Soil organic matter and available water capacity. Journal of Soil and Water Conservation. 49: 189-194. IPCC. 2007. Climate Change 2007: Synthesis Report. Fourth assessment report of the Intergovernmental Panel on Climate Change. Jastrow, J.D. 1987. Changes in soil aggregation associated with tallgrass prairie restoration. American Journal of Botany 74: 1656-1664. Jastrow, J.D. 1996. Soil aggregate formation and the accrual of particulate and mineral- associated organic matter. Soil Biology and Biochemistry 28: 665-676. Jenkinson, D.S., Potts, J.M., Perry, J.N., Barnett, V., Coleman, K., and Johnston, A.E. 1994. Trends in herbage yields over the last century on the Rothamsted long-term continuous hay experiment. Journal of Agricultural Science 122: 365-374. Jenkinson, D.S., Poulton, P.R., Johnston, A.E., and Powlson, D.S. 2004. Turnover of nitrogen-15 labeled fertilizer in old grassland. Soil Science Society of America Journal 68: 865-875. JMP, Version 7, SAS Institute Inc., Cary, NC, 1989-2007. Kemper, W.D., Rosenau, R.C., 1986. Aggregate stability and size distribution. In: Klute, A. (Ed.), Methods of Soil Analysis Part 1- Physical and Mineralogical Methods. Soil Science Society of America, Inc., Madison, WI, USA, pp. 425- 442. Klute, A., 1986. Water retention: laboratory methods. In: Klute, A. (Ed.), Methods of

72 Soil Analysis Part 1- Physical and Mineralogical Methods. Soil Science Society of America, Inc., Madison, WI, USA, pp. 635-662. Knapp, A.K. and T.R. Seastedt. 1998. Introduction: Grasslands, Konza Prairie and Long Term Ecological Research. In: Knapp, A.K., Briggs, J.M., Hartnett, D.C., and Collins, S.L. (Eds.). Grassland Dynamics. Oxford University Press, New York, NY, pp. 3-18. Kucharik, C.J. Fayram, N.J. and Cahill, K.N. 2006. A paired study of prairie carbon stocks, fluxes, and phenology: Comparing the worlds oldest prairie restoration with an adjacent remnant. Global Change Biology. 12: 122-139. Kucera, C.L. and Dahlman, R.C. 1968. Root-rhizome relationships in fire-treated stand of big bluestem, Andropogon gerardii. American Midland Naturalist 80: 268- 271. Lal, R. 1994. Methods and guidelines for assessing sustainable use of soil and water resources in the tropics. Soil Management and Support Service, United Stated Department of Agricultlure. SMSS Technical Monograph no. 21. Lal, R. 1997. Degradation and resilience of soils. Philosophical Transactions of the Royal Society 352: 997-1010. Lal, R. 2007. Soil Science and the Carbon Civilization. Soil Science Society of America Journal 71: 1425-1437. Lal, R. 2008. Carbon sequestration. Philosophical Transactions of the Royal Society 363: 815-830. Lal, R., Kimble, J.M., Follett, R.F. and Cole, C.V. 1999. The Potential of U.S. Cropland to Sequester Carbon and Mitigate the Greenhouse Effect. CRC Press. Boca Raton, FL. Lal, R., Kimble, J.M., Ilvari, T., Sobecki, T.M. 2003. Soil Degradation in the United States. CRC Press, Boca Raton, FL. Lal, R., and Shukla, M. K. 2004. Principles of Soil Physics. Marcel Dekker, Inc., New York, N.Y. Matamala, R., Jastrow, J.D., Miller, R.M., and Garten, C.T. 2008. Temporal changes in C and N stocks and Restored Prairie: Implications for C sequestration strategies. Ecological Applications 18: 1470-1488. McLauchlan, K.K., Hobbie, S.E., and Post, W.M. 2006. Conversion from agriculture to grasslands builds soil organic matter on decadal timescales. Ecological Applications 16: 143-153. Miller, R. M. and Jastrow, J. D. 2000. Mycorrhizal fungi influence soil structure. In: Kapulnik, Y. and Douds, D. D. (Eds.). Arbuscular Mycorrhizae: Physiology and Function. Kluwer Academic Publishers, the Netherlands, pp. 3-18. Nelson, D.W., Sommers, L.E., 1996. Total carbon, organic carbon, and organic matter. In Sparks, D.L. et al. (Ed.), Methods of Soil Analysis, Part 3, ASA and SSSA, Madison, WI, pp.417-435. Pineiro, G., Jobbagy, E.G., Baker, J., Murray, B.C., and Jackson, R.B. 2009. Set-asides can be a better climate investment than corn ethanol. Ecological Applications 19: 277-282. Post, W.M., and Kwon, K.C. 2000. Soil carbon sequestration and land-use change: processes and potential. Global Change Biology 6: 317-327. 73 Rice, C.W., Todd, T.C., Blair, J.M., Seastedt, T.R., Ramundo, R.A. and Wilson, G.W.T. 1998. Belowground Biology and Processes. In: Knapp, A.K., Briggs, J.M., Hartnett, D.C., and Collins, S.L. (Eds.). Grassland Dynamics. Oxford University Press, New York, NY, pp. 244-264. Robertson, G.P., et al. 2008. Sustainable Biofuels Redux. Science 322: 49-50. Schmer, M.R., Vogel, K.P., Mitchell, R.B., and Perrin, R.K., 2008. Net energy of cellulosic ethanol from switchgrass. Proceedings of the National Academy of Sciences U.S.A. 105: 464-469. Seastedt, T.R., and Knapp, A.K. 1993. Consequences of non-equilibrium resource availability across multiple time scales: the transient maxima hypothesis. American Naturalist 141: 621-633. Sollins, P., Glassman, C., Paul, E.A., Swanston, C., Ljtha, K., Heil, J.W., Elliot, E.T., 1999. Soil carbon and nitrogen fractions. In: Robertson, G.P., Coleman, D.C., Bledsoe, C.S., Sollins, P. (Eds.), Standard Soil Methods for Long-term Ecological Research. Oxford University Press, New York, NY, pp. 89-105. Tilman, D., Hill, J. and Lehman, C. 2006. Carbon-negative biofuels from low-input high-diversity grassland biomass. Science 314: 1598-1600. Wander, M.M., and Bollero, G.A. 1999. Soil Quality Assessment of Tillage Impacts in Illinois. Soil Science Society of America Journal 63: 961-971. Wilson, G. W. T., Rice, C. W., Rillig, M. C., Springer, A. and Hartnett, D. C. 2009. Soil aggregation and carbon sequestration are tightly correlated with the abundance of arbuscular mycorrhizal fungi: results from long-term field experiments. Ecology Letters 12: 1-10. Yoder, R.E. 1936. A direct method of aggregate analysis of soils and a study of the physical nature of erosion losses. Journal of the American Society of Agronomy 28: 337-351.

74 CHAPTER 3

THE LONG-TERM EFFECTS OF THE CONVERSION FROM

TALLGRASS PRAIRIE TO WHEAT PRODUCTION ON

SOIL ORGANIC CARBON IN NORTH CENTRAL KANSAS

3.1 Abstract

Land use change, cultivation, soil management, and plant community composition directly affect the level and dynamics of soil organic carbon (SOC). This study describes and quantifies the long-term effects the conversion of a perennial plant community to an annual plant-based agricultural system on SOC pools in sites converted from tallgrass prairie to annual agriculture. The long term effects of land use change on SOC pools are analyzed by sampling five farms that contain both annually harvested tallgrass prairie remnants (PM) and conventionally farmed wheat (Triticum aestivum) fields (AG) on the same soil types.

Soil core samples were collected to a depth of 1m in May and June of 2008.

Management effects on SOC pools were assessed by analyzing total soil organic C

(SOC), total soil nitrogen (TSN), microbial biomass C (MBC) and a particle size fractionation of SOC in coarse sand (>250µm), fine sand (250-53µm), silt (53-2µm), and clay (<2µm) sized fractions. PM soils showed statistically higher levels of all parameters 75 measured to a depth of 60cm. SOC pools were decreased by 30% in the AG soils (59 Mg

C ha-1) from PM soils (84 Mg C ha-1). PM soils had an average of 7.71 Mg N ha-1 in the

0-40cm depth, while AG soils contained 5.54 Mg N ha-1 at these depths; a 28% reduction.

PM soils had four times as much MBC in the soil surface as AG fields, 257 µg C g-1 soil compared to 64 µg C g-1 soil. PM soils had increased SOC levels in all particle size fractions. Clay-sized particles were observed as the dominant fraction of SOC in both soils. These data support observations that perennial plant communities, such as the prairie meadows, store and cycle SOC at far greater levels than annual plant communities.

3.2 Introduction The conversion of native to agricultural ecosystems, based on annual plants, has been among the most significant human interventions into global landscapes and biogeochemical cycles. The resulting ecosystem changes that occur often lead to environmental pollution and land degradation, and to agriculture being labeled as threat to healthy natural systems. Biomass alterations, tillage, fertilization and altered hydrology are four of the primary vectors through which agriculture affects ecosystem processes (McLauchlan 2006). In many cases an almost predictable pattern has emerged from scenarios of converting natural landscapes to agriculture. As forests and grasslands are cleared and cultivated, the soil is exposed to the elements, and the extensive perennial root systems, that support the functioning of soil physical and biological health, are lost. Massive soil erosion events tend to follow the conversion to agriculture and are a common feature in the history of societies around the world

76 (Montgomery 2007). A slow, generally unnoticed, drawdown of the remaining soil resource then ensues, as continued cultivation and residue removal, and the concurrent soil degradation, lead to declining yields and agricultural collapse (Montgomery 2007,

Lal 2007). The cycle continues as humans are forced to continue converting new land to agriculture to meet their needs (Montgomery 2007; Lal 2007).

A significant factor in the soil degradation following the conversion of natural systems to agriculture is the loss of soil organic C (SOC) that occurs when soil cultivation is initiated. A recent global analysis of 74 studies of land conversions found that forests and grasslands lost 42% and 59%, on average, of their SOC pool when they were converted to cropland (Guo and Gifford 2002). Another review, focused primarily in temperate regions, concluded that an average of 20-40% of SOC is lost within 5-20 years of the onset of cultivation (Davidson and Ackerman 1993). A comparative study on SOC and soil cultivation in different climatic and biotic regions indicated that roughly half of the antecedent SOC in a temperate grassland can be lost after 50 years of cultivation, if sufficient inputs are not added to the system; the same level of SOC loss can occur in less than 10 years in the tropics (Tiessen et al. 1994). A significant amount of the soil’s inorganic nutrients are associated with SOC and lost in this process (Tiessen et al. 1994), adding to the stress placed on the soil and the agricultural system. This ensures that increasing levels of inputs will be required to maintain production levels.

The SOC lost in land uses change is also a globally significant factor in the C cycle and makes up no less than 15% of annual anthropogenic C emissions (IPCC 2007).

The onset of cultivation is known to change the storage and cycling of SOC in a number of ways. The soil aggregate structure is broken down and reduced by cultivation 77 (Gupta and Germida 1988, Cambardella and Elliott 1993). This exposes organic matter, previously occluded in aggregates, to decomposition. Soil structural degradation, and the concurrent reduction of pore space, further affects the soil by reducing soil microbial activity and restricting plant root growth. Cultivation is also known to cause rapid mineralization of particulate organic matter (POM)(Cambardella and Elliot 1992). POM is made up primarily of partially decomposed plant residues and is an important substrate for biological activity and an important binding agent for aggregation. The disturbance of cultivation reduces the microbial biomass C (MBC) pool (Smith and Paul 1990), an important component of C and N cycling in the soil. In addition to the effects of cultivation on the soil system, the SOC balance is also altered by reduced inputs and increased decomposition, in many soils converted to agricultural systems (Huggins et al.

1998).

Growing awareness of the problems of soil and resource degradation associated with agriculture led to strong movements towards soil conservation (Bennett 1947) and ecological agriculture (Howard 1940) in the 20th century. Recognizing the role of tillage, or cultivation, in the processes associated with soil erosion and degradation, agronomists have developed productive no-till agricultural systems. These systems have proven highly effective at maintaining high levels of productivity while enhancing soil quality and conserving the soil resource (Blevins et al. 1998). Replacing the, relatively recent, use of chemical fertilizer inputs with agronomic systems based on organic inputs

(Reganold et al. 1987), such as compost and manure, and more complex crop rotations involving cover crops (Drinkwater et al. 1998) have also led to enhanced soil quality and conservation in agricultural systems. Recently an emphasis has also been placed on 78 managing agricultural soils to sequester C, thereby improving the soil and adding to the offset of C emissions (Lal 2004).

Others have noted that while these advances have improved the ecology of agricultural systems, those systems are still quite inefficient, compared to the ecosystems they replaced. They argue that the single biggest factor contributing to the environmental degradation and energy consumption of agriculture is the annual life cycle of most cultivated plants (Cox et al. 2006, Jackson 2002). Frequent tillage and heavy use of fertilizers and herbicides are all strategies for meeting the needs of annual plant communities, which have just one season to complete their life cycle. Perennial crops offer several distinct physiological and ecological advantages over their annual counterparts. Perennials tend to have a deeper root system and a longer growing season than their annual counterparts (Glover et al. 2007). Their extensive root systems make them more efficient users of water and nutrients, and allow them to sequester more C in the soil than annual crops (Cox et al. 2006, Glover et al. 2007).

Long term ecological studies in which annual and perennial based plant communities have been compared have found that perennials maintain superior levels of ecosystem functioning. In a study looking at erosion over 100 years at the Sanborn field in Missouri, plots in continuous corn (Zea mays) cultivation retained only 44% of the topsoil found in continuous timothy grass (Phleum pratense) hay plots (Gantzer et al.

1991), and soil erosion rates under continuous corn were estimated to be 50 times greater than erosion rates under the timothy grass (Gantzer et al. 1990). Soil sampling at long- term field trials at the Kellogg Biological Station in Michigan indicated that a conventionally grown corn-soybean-wheat rotation was not gaining or losing SOC, while 79 an organic version of the same system was accumulating 12.3 g C m-2 yr-1 (Grandy and

Robertson 2007). Perennial treatments at the station showed much higher rates of SOC accumulation with alfalfa increasing SOC at 28 g C m-2 yr-1 and early successional hardwood forest gaining 31.6 g C m-2 yr-1 (Grandy and Roberson 2007).

The increased accumulation of SOC is likely due to the increased allocation of C to belowground pools in perennial plant communities. This difference in C allocation between perennials and annuals can be seen at the level of individual plants and at the ecosystem level. A comparison of photosynthetic C allocation in two species of Bromus grass, one annual and one perennial, found that the perennial species had greater annual allocations of C to belowground pools (Warembourg and Esterlich 2001). A study comparing the ecosystem C budgets of winter wheat (Triticum aestivum) and tallgrass prairie indicated that, while the wheat had a higher annual allocation of NPP to belowground biomass, the prairie system had an almost tenfold increase in belowground residue accumulation annually (Buyanovsky et al. 1987). That study also indicated that the prarie was taking up water from the soil for a much greater period during the growing season (Buyanovsky et al. 1987). This limited the decomposition belowground C in the prairie, compared with the wheat (Buyanovsky et al. 1987). The quantity and quality of

SOC are known to be a central to numerous ecosystem processes (Lal 2007b), and it may be that the greater allocation of C in perennial plant communities is a key feature of maintaining ecosystem services in those systems.

High levels of SOC and ecosystem services have generated significant research interest in the possibility of agricultural systems based on perennial grasslands, such as the tallgrass prairie. Prairie grasses have also shown great promise as biofuel crops by 80 being efficient producers of cellulosic ethanol (Schmer et al. 2008) and offering C negative conversion for croplands (Fargione et al. 2008). Prairie-like low-input, high diversity grassland systems were able to produce more biomass and soil C per unit area than grass monocultures, while receiving no inputs (Tilman et al. 2006). Researchers at the Land Institute are in the process of developing perennial wheat, and other grains, which will produce quantities of grain similar to modern hybrid varieties while offering the ecological benefits of perennial grasses (Cox et al. 2006). Long-term studies validate the sustainability of agriculture based on perennial grasses. Neither total soil nitrogen

(TSN) (Jenkinson et al. 2004) nor biomass yields (Jenkinson et al. 1994) have declined in unfertilized grasslands after 120 years of twice annual hay harvests, at Rothamsted, U.K.

This study compares SOC fractions between unfertilized tallgrass prairie meadows harvested annually for hay and adjacent fields, on similar soil types, annually cropped in winter wheat. Glover et al. (2009) reported that after 75 years of continuous hay harvesting, the prairie meadows produced greater annual biomass yields and a 23% more cumulative N yield ha-1, while receiving only 8% of the energy inputs of the wheat system. The prairie meadows supported these high levels of productivity while also maintaining a much higher level of ecosystem services including: higher levels of insect diversity, increased levels of total soil nitrogen, and reduced riverine nitrate levels in the surrounding watersheds (Glover et al. 2009). Soil nematode communities in the prairie meadows suggested increased soil food web complexity and stability, compared with the wheat fields (Culman et al 2009).

The objective of this study was to examine changes in SOC fractions between the prairie meadow and wheat ecosystems. Total SOC, TSN, and MBC pools were analyzed 81 and compared. A particle size fractionation technique was used to examine SOC in the primary particle size classes of sand, silt and clay. These indices were observed in support of the primary objective of the study: To determine changes in SOC storage in cycling in annually cropped wheat fields, compared with annually harvested remnant tallgrass prairie meadows. The hypothesis tested was that: the SOC levels have declined significantly in the conversion of prairie meadows to wheat production.

3.3 Materials and Methods 3.3.1 Study Site This study took place on five farms, located in five counties in north central

Kansas, which all contain a similar, unique pairing of land uses, on bottomland soils.

These farms all contain remnant tallgrass prairie meadows, preserved for their utility as hay fields, and conventionally farmed winter wheat (Triticum aestivum) fields. These sites all occur on soils that are classified as Prime Farmland in the USDA Natural

Resource Conservation Services agricultural land classification. Remnant prairies are extremely rare in this region, on soils that are well suited to agriculture. These sites have been preserved in prairie either due to irregular or small shape, which makes annual agriculture difficult, (the Five Creek, Buckeye, and Niles sites) or because of long standing family traditions (the New Cambria and Goessel sites) (Glover et al 2009).

The site names, county names and geographic locations of the farms are the following: Niles, Ottawa Co. N 38°58’145”, W 97°28’616”; Buckeye, Dickinson Co. N

39°2’344”, W 97°7’798”; New Cambria, Saline Co. N 38°53’54”, W 97°32’615; Five

Creek, Clay Co. N 38°22’665”, W 97°18’788”; Goessel, McPherson Co. N 38°15’333”,

82 W 97° 22’307”. The soils are comprised of soil series from 3 different soil orders (Table

3.1).

Site Soil Series Soil Taxonomy

Niles Hord Fine-silty, mixed, superactive, mesic,

Cumulic Haplustoll

Buckeye Hobbs Fine-silty, mixed, superactive, nonacid, mesic,

Mollic Ustifluvent

Five Creek Muir Fine-silty, mixed, superactive, mesic,

Cumulic Haplustoll

New Cambria Detroit Fine, smectitic, mesic, Pachic Argiustoll

Goessel Goessel Fine, smectitic, mesic, Typic Haplustert

Table 3.1: Soil taxonomic descriptions of research sites.

Previous studies of the vegetation in the prairie meadows indicates that they share a similar plant community composition with an average of 79% of the ground cover made up of the following plant functional groups, native to the tallgrass prairie region: perennial grasses (69%), legumes (7%), and non-legume forbs (3%) (Glover et al 2009).

An additional 4% of ground cover was composed of non-native annual grasses (Glover et al 2009). The vegetation in these meadows is clipped to 8-10cm each year, during mid- summer, dried as hay and used for livestock fodder (Glover et al 2009). Landowners all reported that the meadows have never received fertilizer application. Both the wheat 83 production and the harvesting of the hay meadows are known to have been in practice on theses farms for approximately 75 years. The mean annual temperature in the region is

13.2 °C and the mean annual precipitation is 730mm.

3.3.2 Soil Sampling Soil sampling was carried out in mid May (peak wheat biomass) and late June

(peak prairie biomass) 2008. Five 4cm diameter soil cores were taken in each treatment field to a depth of 1m. Cores were divided into 0-10, 10-20, 20-40, 40-60, 60-80, and 80-

100cm depths. The five cores for each sampling block were then bulked according to depth and mixed until homogenous. The mass of cores was recorded for each depth for

bulk density (ρb) measurements. 100g sub-samples were taken from each sample and dried for 24h at 105°C to determine the gravimetric moisture content (ω).

3.3.3 Total C and N Concentrations Air-dried soil, from the June sampling date, was passed through a 2mm sieve and ground with a mortar and pestle in preparation for C/N analysis. Samples in the range of

40-60mg, based on estimated C and N concentrations, were weighed and encased in

5x9mm Costech sampling. These samples were then shipped to the Stable Isotope

Facility at the University of California, Davis, where they were analyzed for total soil C and total soil N (TSN) by combustion, using a PDZ Europa ANCA-GSL elemental analyzer (Sercon Ltd., Chesire, UK). Previous analysis indicated that the pH in these soils was <6.6, so inorganic carbonate composition in these soils is negligible and total C

84 is considered soil organic C (SOC) in this study. SOC and TSN pools were calculated according to the following equation (Eq 1) (Lal et al. 1999):

-1 -3 4 2 -1 Mg C or N ha = %C or N × ρb (Mg m ) × depth (m) × 10 m ha 100 …………..(Eq 1)

3.3.4 Microbial Biomass C Microbial biomass carbon (MBC) was obtained using the chloroform fumigation extraction method (Vance 1987). Briefly, three 10g samples of field moist soil, sieved through at 6.75mm sieve, were taken from each sample. One 10g sample served as a control, one as a fumigated sample, and the third was used for moisture content.

The fumigated samples were placed in a vacuum dessicator with approximately 50ml of chloroform. The dessicator was evacuated until the chloroform bubbled for 1 minute.

The dessicator was then sealed and the samples were left in the dessicator for 24 hours.

Following the fumigation, the samples were extracted by placing them in 250ml plastic

bottles with 80ml of .05M K2(SO4). The bottles were then shaken in the floor shaker for

1 hour. The shaken samples were centrifuged for 5 minutes and the supernatant was passed through at 0.45µm syringe filter. Control samples were extracted using the same process, but were not fumigated. Two analytical replicates were analyzed for each sampling date, and blanks were prepared with each replicate.

Filtered samples were frozen in 50ml centrifuge tubes. Frozen samples were sent to the Stable Isotope Facility at the University of California at Davis, where they were analyzed for total organic carbon (TOC) with an O.I. Analytical Model 1010 TOC

Analyzer (OI Analytical, College Station, TX).

85 The MBC was calculated using equations from the method of Voroney et al. (1993).

All soil weights were corrected for moisture content. Total weight of extractable C in the fumigated (OF) and unfumigated (OUF) soil samples:

-1 -1 OCF, OCUF (µg g soil ) = extractable C (ug mL ) X VS (ml) / MS (g)

(Eq 2)

Where VS is the volume of the extractant solution and MS is the mass of soil in the solution.

Microbial Biomass C in the soil (MBC):

-1 MBC (ug g soil ) = (OCF – OCUF)…………………………………(Eq 3)

3.3.5 Particle Size Fractionation A primary particle size fractionation (Christensen 1992) was carried out to divide the soil into coarse sand (2mm-250µm), fine sand (250-53µm), silt (53-2µm), and clay

(<2µm) sized fractions. The soil was dispersed chemically, with Na-HMP, using a method outlined in Cambardella and Elliot (1992) and Sollins et al. (1999). Air-dried soil, from the June sampling date, was passed through a 2mm sieve. 20g of soil were placed in a 250ml plastic bottle in 60ml of 5g L-1 Na-HMP and shaken overnight (16h) in a floor shaker. The resulting solution was then passed through nested 250µm and 53µm sieves. DI water was used to wash the solution through the sieves and the material captured on the sieves was washed into pre-weighed glass beakers and oven dried

(<45°C). The resulting material was considered as the sand sized fractions and POM.

The soil slurry that passed through the sieves was collected in plastic cylinders and separated into silt and clay by sequential decantation and siphoning (Rutledge et al. 86 1967). The silt remaining in the cylinders was washed into pre-weighed glass beakers

and oven dried. The clay suspension was flocculated using 0.5M MgCl2 prior to oven drying. The coarse sand (CS), fine sand (FS), and silt (SI) were weighed to calculate their relative composition to the whole soil, and the mass of the clay (CL) fraction was obtained by subtraction.

Fractionated samples were ground with a mortar and pestle, and 40-60mg samples were prepared, based on estimated C and N concentration, for C/N analysis. Samples were packed in Costech 5×9mm sample tins and shipped to the Stable Isotope Facility at the University of California, Davis where they were analyzed for total C and total N by combustion using a PDZ Europa ANCA-GSL (Sercon Ltd., Chesire, UK).

C and N content in each fraction were calculated (in g C or N kg soil-1) using the following equation from Sollins et al (1999):

C(fr) = Cfr x Wfr x 10……………….………………………………..(Eq 4)

Where the C, or N, in each fraction, C(fr), is equal to the % concentration of C in that

fraction (Cfr) multiplied by the dry mass of that fraction (Wfr) (g fraction/g soil) multiplied by ten to achieve C in g kg soil-1.

3.3.6 Data Analysis All data was analyzed using analysis of variance (ANOVA) using the PROC MIXED procedure in SAS software (Cary, N.C.). Depth and treatment were treated as fixed effects, and site was treated as a random effect, in the analysis. Statistical significance was determined at α = 0.05 level.

87 3.4 Results and Discussion 3.4.1 Soil organic C The Prairie Meadow (PM) treatment contained significantly higher levels of SOC concentration than the wheat production (AG) treatment from the soil surface to a depth of 60cm (Figure 3.1). This pattern of SOC loss was also reflected in the values measured for the SOC pools (Table 3.2 and Figure 3.4) The PM sites contained 84 Mg C ha-1 in the surface 40cm of the profile and 153 Mg C ha-1 from the soil surface to 1m depth. The

AG sites contained 59 Mg C ha-1 in the upper 40cm of the soil and 115 Mg C ha-1 to a depth of 1m. This represents a 30% loss of SOC in the upper 40cm of the profile and

24% loss of SOC if the 1m profile is considered. This level of SOC loss is in accord with the observations reviewed in the Davidson and Ackerman (1993) study which reported an average of 20-40% SOC lost after the conversion of natural systems to agriculture. Several of the studies included in this review were taken from soils in the prairie region of the Midwestern U.S. These observations suggest a potential C sink of

25 Mg C ha-1 in the upper 40cm of these soils.

The SOC concentration showed similar patterns in the depth profiles of the prairie meadows (Fig 3.2), across soil types. SOC concentration varied between 2-3% in the surface layer in the prairie meadows and 1.5-0.5% below 40cm. The pattern of SOC loss was not uniform across the depth profiles of the wheat fields from the separate sites (Fig

3.3), but all sites showed marked decreases in SOC through 40cm depth.

88 Soil Organic C (%) 0 0.5 1 1.5 2 2.5

10 ***

20 ***

40 *** Prairie Meadows Wheat Fields 60 *** Depth (cm)

80

100

Figure 3.1: Depth Profile of Soil Organic C Concentration (%) in long term study sites. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Triple asterisk indicates mean values are significantly different (p <.0001). Bars indicate the standard error.

89 SOC Conc (%) 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 0

20

40 Niles Buckeye 60 Gossel New Cambria Five Creek

Depth (cm) 80

100

120

Figure 3.2: Depth Profile of SOC Concentration in the Prairie Meadows of the Five Long-term study sites.

90 SOC Conc (%) 0.00 0.50 1.00 1.50 2.00 0

20

40 Niles Buckeye 60 Goessel New Cambria Five Creek

Depth (cm) 80

100

120

Figure 3.3: Depth Profile of SOC Concentration in the Wheat Fields of the Five Long-term study sites.

91

Soil Organic C Pools (Mg C ha-1)

Depth (cm) Prairie Meadows Wheat Fields S.E.

0-10 **29.29 19.85 ±2.21

10-20 **28.37 19.69 ±1.89

20-40 **26.53 19.36 ±1.43

40-60 **24.68 19.04 ±1.38

60-80 22.83 18.71 ±1.79

80-100 20.99 18.38 ±2.44

Table 3.2: Depth Distribution of SOC pools (Mg C ha-1) in long term study sites. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Double asterisk indicates mean values are significantly different (p <.01).

92 SOC Pools (Mg/ha) 0 5 10 15 20 25 30 35

** 10

20 **

40 ** Prairie Meadows Wheat Fields 60 ** Depth (cm)

80

100

Figure 3.4: Depth Profile of SOC Pools (Mg C ha-1) in long term study sites. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Double asterisk indicates mean values are significantly different (p <.01). Bars indicate the standard error.

93 3.4.2 Total soil N The PM sites had significantly larger TSN pools (Table 3.3) and TSN concentration

(Fig 3.5) than the AG sites to a depth of 60cm. These figures follow similar trends as the

SOC pools (Table 3.2) and SOC concentrations (Fig 3.1). The PM sites had an average of 7.71 Mg N ha-1 in the upper 40cm of the profile and 14.26 Mg N ha-1 to a depth of 1m.

The AG sites had 5.54 Mg N ha-1 in the upper 40cm and 11 Mg N ha-1 in the top 1m of the soil. This difference in TSN between the PM soils and the AG soils represents a loss of 28% of TSN in the upper 40cm and a 23% loss of TSN in the m profile.

The significant loss of TSN in the AG soils seems to indicate at least two differences between these systems. The N in these soils appears to have a strong correlation with the SOC. The loss of SOC from cultivation and management of the winter wheat system has decreased the AG soils ability to hold N. Secondly, the perennial plant community appears to be far more efficient at storing and cycling N.

These results are especially striking given the management history of the sites. Both systems have exported an average of 47 kg N ha-1 yr-1, in the form of the hay and grain being harvested (Glover et al. 2009). The wheat system has received 70 kg N ha-1 yr-1 in fertilizer inputs, while the meadows have never been fertilized (Glover et al. 2009). The fact that the prairies still contain an additional 3.26 Mg N ha-1 in the soil profile, despite the discrepancy in inputs, is strong evidence that the prairie system cycles N efficiently than the wheat systems.

94 Total Soil N (%) 0 0.05 0.1 0.15 0.2 0.25

*** 10

20 ***

40 *** Prairie Meadows Wheat Fields 60 *** Depth (cm)

80

100

Figure 3.5: Depth Profile of Total Soil N Concentration (%) in long term study sites. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Triple asterisk indicates mean values are significantly different (p <.0001). Bars indicate the standard error.

95

Total Soil Nitrogen Pools (Mg N ha-1)

Depth (cm) Prairie Meadows Wheat Fields S.E.

0-10 **2.66 1.85 ±0.19

10-20 **2.59 1.85 ±0.16

20-40 **2.46 1.84 ±0.12

40-60 **2.32 1.83 ±0.12

60-80 2.18 1.82 ±0.16

80-100 2.05 1.81 ±0.21

Table 3.3: Depth Distribution of TSN pools (Mg N ha-1) in long term study sites. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Double asterisk indicates mean values are significantly different (p <.01).

96 Total Soil N Pool (Mg/ha) 0 0.5 1 1.5 2 2.5 3

** 10

20 **

40 ** Prairie Meadows Wheat Fields ** 60 Depth (cm)

80

100

Figure 3.6: Depth Profile of TSN Pools (Mg N ha-1) in long term study sites. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Double asterisk indicates mean values are significantly different (p <.01). Bars indicate the standard error.

97 3.4.2 Microbial Biomass C The MBC analysis measured significantly higher levels of MBC in the PM treatment through 80cm depth, in values averaged across both the May and June sampling dates.

The PM had four times as much MBC as the AG treatment in the soil surface layer through a depth of 40cm, three times as much MBC at 60cm and twice as much MBC at

80cm depth. The MBC concentration in the extraction represented just over 1% of the total SOC in the prairie meadows. These values are within the range of values observed in literature values observed in studies measuring MBC in temperate grasslands and agricultural systems. The values measure in the layers close to the surface (0-40cm) of the prairie (PM) treatment are very close to the 217µg C g soil-1 MBC value recorded for

0-30cm in Garcia and Rice (1994), a study which also took place in tallgrass prairies in central Kansas.

Values recorded are presented as µg C g-1 soil measured in the MBC extracts. These

values were not converted to MBC values with KEC conversion value, which estimates the

efficiency with which the MBC is extracted. This decision was made because a KEC value taken from the literature is not responsive to specific soil and climatic conditions of the site, or with dynamics that change with increasing depth in the soil profile. Typical

KEC values would keep the MBC in the expected range of 1-3% of SOC.

98 MBC (ug/g) 0 50 100 150 200 250 300

10 ***

20 ***

Prairie Meadows 40 *** Wheat Fields

Depth (cm) *** 60

** 80

Figure 3.7: Depth Profile of Microbial biomass C (µg C g-1 soil) measured in long term agriculture study sites. Values are means from ANOVA run in PROC MIXED, SAS v9.1. Asterisk indicates mean values are significantly different: triple asterisk (p <.0001) and double asterisk (p <.01). Bars indicate the standard error.

99

Microbial Biomass Carbon (µg C g-1 soil)

Depth (cm) Prairie Meadows Wheat Fields S.E.

0-10 ***256.87 64.36 ±8.90

10-20 ***227.15 58.21 ±7.62

20-40 ***167.72 45.91 ±5.75

40-60 ***108.29 33.61 ±5.59

60-80 **48.86 21.31 ±7.23

Table 3.4: Depth Distribution of MBC (µg C g-1 soil) in long term study sites. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Asterisk indicates mean values are significantly different: triple asterisk (p <.0001) and double asterisk (p <.01).

100 3.4.5 Particle Size Fractionation

3.4.5.1 The Coarse Sand Fraction Significant decreases in SOC were observed to 60cm depth in the coarse sand (>250µm) fraction (Table 3.5). The coarse sand SOC showed a 48% decrease in the top 10cm and a

49% decrease to 60cm depth (0-60cm). This fraction, along with the fine sand fraction, is considered part of the particulate organic matter (POM). Several previous studies have indicated that SOC in the POM fractions undergoes heavy losses and rapid mineralization in soils under long-term cultivation (Tiessen and Stewart 1983, Cambardella and Elliot

1992, Wander and Bollero 1999).

Suprisingly, the N in the coarse sand fraction did not demonstrate statistically significant decreases in the long-term wheat fields (Fig 3.8). This is the only fraction, of the 4 measured, that did not have large losses of TSN coinciding with the SOC losses. It may be that this fraction has been enriched in N through the application of inorganic N fertilizers on the wheat fields. Another possibility is that this fraction is very N rich in the wheat fields.

101

SOC in the Coarse Sand (>250µm) Fraction (g C kg-1 soil)

Depth (cm) Prairie Meadows Wheat Fields S.E.

0-10 **0.97 0.50 ±0.11

10-20 **0.86 0.44 ±0.09

20-40 **0.63 0.32 ±0.07

40-60 *0.39 0.20 ±0.07

60-80 0.16 0.08 ±0.12

Table 3.5: Depth Distribution of SOC in the Coarse Sand Fraction (g C kg-1 soil) in long term study sites. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Asterisk indicates mean values are significantly different: double asterisk (p<.01) and single asterisk (p <.05).

102 SOC in the >250um Fraction (g C / kg soil) 0 0.2 0.4 0.6 0.8 1 1.2

** 10

** 20

** Prairie Meadows 40 Wheat Fields

Depth (cm) * 60

80

Figure 3.8: Depth Profile of SOC in the Coarse Sand Fraction (g C kg-1 soil) in long term study sites. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Asterisk indicates mean values are significantly different: double asterisk (p<.01) and single asterisk (p <.05). Bars indicate the standard error.

103 TSN in the >250um Fraction (g N / kg soil) 0 0.01 0.02 0.03 0.04 0.05

10

20

Prairie Meadows 40 Wheat Fields Depth (cm) 60

80

Figure 3.9: Depth Profile of TSN in the Coarse Sand Fraction (g N kg-1 soil) in long term study sites. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Bars indicate the standard error.

104 3.4.5.2 The Fine Sand Fraction The PM sites contained significantly larger quantities of SOC in the fine sand fraction (250-53µm) than the AG sites through 60cm depth (Table 3.6). This trend is similar to the one observed in the total SOC pools (Fig 3.4). The PM sites contained roughly twice as much SOC in this fraction, as the AG sites, through the 60cm depth, indicating that a sizeable portion of this fraction has been mineralized during the 75 years of wheat production. This fraction showed a 54% reduction in SOC in the upper 10cm of the soil and a 52% reduction when 0-60cm depth was considered. This trend is consistent with previous research literature that demonstrated that SOC, in the form of

POM, in the sand size fractions is among the first SOC to be mineralized when temperate grasslands are cultivated (Cambardella and Elliot 1992). Below 60cm, management does not appear to have a strong impact on this fraction, as both treatments have low levels;

PM (0.26 g C kg-1 soil) and AG (0.20 g C kg-1 soil). POM, in sand fractions, has shown strong correlations with soil quality in temperate agricultural soils (Wander and Bollero

1999), and the sizeable loss observed in the AG sites represents a sign of the degradation of the soil ecosystem resulting from long term agricultural production.

N in the fine sand fraction was observed in significantly higher quantities in the PM sites to 40cm depth. The PM soils contained approximately twice as much N in the fine sand fraction as the AG soils through 40cm depth. POM has been strongly correlated with organic N mineralization (Wilson et al. 2001) and the superior quantity of N in the fine sand fraction of the PM soils suggests that the prairie system stores N more effectively than the wheat system.

105

SOC in the Fine Sand (250-53µm) Fraction (g C kg-1 soil)

Depth (cm) Prairie Meadows Wheat Fields S.E.

0-10 ***2.20 1.01 ±0.19

10-20 ***1.92 0.89 ±0.17

20-40 **1.37 0.66 ±0.13

40-60 *0.81 0.43 ±0.12

60-80 0.26 0.20 ±0.16

Table 3.6: Depth Distribution of SOC in the Fine Sand Fraction (g C kg-1 soil) in long term study sites. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Asterisk indicates mean values are significantly different: triple asterisk (p<.0001), double asterisk (p<.001) and single asterisk (p <.05).

106 SOC (g C/kg soil) 0 0.5 1 1.5 2 2.5 3

10 ***

*** 20

** Prairie Meadows 40 Wheat Fields Depth (cm) * 60

80

Figure 3.10: Depth Profile of SOC in the Fine Sand (250-53µm) Fraction (g C kg-1 soil) measured in long term agriculture study sites. Values are means from ANOVA run in PROC MIXED, SAS v9.1. Asterisk indicates mean values are significantly different: triple asterisk (p<.0001), double asterisk (p<.001) and single asterisk (p <.05). Bars indicate the standard error.

107 N (g N/kg soil) 0 0.05 0.1 0.15 0.2

10 ***

*** 20

** Prairie Meadows 40 Wheat Fields Depth (cm) 60

80

Figure 3.11: Depth Profile of TSN in the Fine Sand (250-53µm) Fraction (g N kg-1 soil) measured in long term agriculture study sites. Values are means from ANOVA run in PROC MIXED, SAS v9.1. Asterisk indicates mean values are significantly different: triple asterisk (p <.001) and double asterisk (p <.01). Bars indicate the standard error.

108 3.4.5.3 The Silt Fraction The SOC in the silt-sized (53-2µm) fractions showed significant decreases in the wheat fields to a depth of 60cm. Approximately 57% of the SOC was lost in the silt fraction in the soil surface layer (0-10cm) and 51% of the SOC in the silt fraction was lost when 0-60cm depth is considered. Silt and clay fractions of SOC have been described as having slow turnover and relatively long residence time in the soil (Lutzow et al. 2007), and losses in these fractions may be indicative of the long-term effects of cultivation on relatively stable fractions of C.

Similarly, large levels of N loss were also observed in the silt fraction, to a depth of 60cm.

109

SOC in the Silt (53-2µm) Fraction (g C kg-1 soil)

Depth (cm) Prairie Meadows Wheat Fields S.E.

0-10 ***6.68 2.89 ±0.36

10-20 ***5.94 2.69 ±0.31

20-40 ***4.46 2.28 ±0.23

40-60 **2.98 1.86 ±0.22

60-80 1.50 1.45 ±0.29

80-100 1.01 1.04 ±0.40

Table 3.7: Depth Distribution of SOC in the Silt (53-2µm) Fraction (g C kg-1 soil) in long term study sites. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Asterisk indicates mean values are significantly different: triple asterisk (p<.0001) and double asterisk (p<.001).

110 SOC in the Silt Fraction (g C / kg soil) 0 2 4 6 8

** 10

** 20

40 ** Prairie Meadow Wheat Fields 60 * Depth (cm)

80

100

Figure 3.12: Depth Profile of SOC in the Silt (53-2µm) Fraction (g C kg-1 soil) measured in long term agriculture study sites. Values are means from ANOVA run in PROC MIXED, SAS v9.1. Asterisk indicates mean values are significantly different: triple asterisk (p<.0001), double asterisk (p<.001) and single asterisk (p <.05). Bars indicate the standard error.

111 TSN in the Silt Fraction (g N / kg soil) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

*** 10

*** 20

40 *** Prairie Meadows Wheat Fields 60 ** Depth (cm)

80

100

Figure 3.13: Depth Profile of TSN in the Silt Fraction (g N kg-1 soil) in long term study sites. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Asterisk indicates mean values are significantly different: Triple asterisk (p<.0001) and double asterisk (p<.001).

112 3.4.5.4 The Clay Fraction Both C and N were significantly decreased in the clay fraction (<2µm) of the long-term wheat production soils (Figs 3.14 and 3.15). SOC in the clay fraction was decreased by 29% in the upper 10cm of the soil, from 13.07 g C kg-1 soil in the prairie meadows to 9.24 g C kg-1 soil in the wheat fields. 26% of the SOC in the clay fraction was lost, when 0-60cm depth is considered. Like the silt fraction, these losses represent the mineralization of more stable portions of the SOC. The losses in the clay fraction are, however, markedly less than the losses observed in both the silt and fine sand fractions, that lost just over 50% of their SOC, and the coarse sand fraction that lost just under

50%. This data appears to demonstrated that C in the clay size fraction is more resistant to the disturbances associated with long-term wheat production in these soils.

TSN also showed significant decreases to a depth of 60cm in the clay fraction

(Fig 3.15). It appears that N follows similar trends to SOC in these soils, and that the clay fraction is the dominant contributor to TSN. Again the clay has demonstrated strong retention properties as the proportion of N lost in the clay fraction is much smaller than the proportion of N lost in the silt and fine sand fractions.

113

SOC in the Clay (<2µm) Fraction (g C kg-1 soil)

Depth (cm) Prairie Meadows Wheat Fields S.E.

0-10 ***13.07 9.24 ±0.63

10-20 ***12.11 8.73 ±0.54

20-40 ***10.20 7.70 ±0.41

40-60 **8.29 6.68 ±0.40

60-80 6.38 5.66 ±0.51

80-100 4.47 4.64 ±0.70

Table 3.8: Depth Distribution of SOC in the Clay Fraction (g C kg-1 soil) in long term study sites. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Asterisk indicates mean values are significantly different: triple asterisk (p<.0001) and double asterisk (p<.001).

114 SOC in the Clay Fraction (g C / kg soil) 0 5 10 15

10 ***

20 ***

*** 40 Prairie Meadows Wheat Fields 60 ** Depth (cm)

80

100

Figure 3.14: Depth Profile of SOC in the Clay Fraction (g C kg-1 soil) in long term study sites. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Asterisk indicates mean values are significantly different: triple asterisk (p<.0001) and double asterisk (p<.001).

115 N in Clay Fraction (g N / kg soil) 0 0.5 1 1.5

*** 10

*** 20

*** 40 Prairie Meadows Wheat Fields 60 ** Depth (cm)

80

100

Figure 3.15: Depth Profile of TSN in the Clay Fraction (g C kg-1 soil) in long term study sites. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Asterisk indicates mean values are significantly different: triple asterisk (p<.0001) and double asterisk (p<.001).

116 3.4.5.5 Relative Composition of SOC by the Particle Size Fractions The graphs below (Fig 3.16 and 3.17) show the relative composition of SOC by the particle size fractions. These graphs clearly demonstrate that the clay fraction is the dominant size of SOC in these soils. Theses figures also clearly indicate that the relative contribution of clay to the total SOC has increased in the wheat fields to the 60cm depth of SOC disturbance.

Relative Composition of Total SOC by Particle Size Fractions (%) 0% 20% 40% 60% 80% 100%

10

20

40 Clay Silt Fine Sand 60 Coarse Sand Depth (cm)

80

100

Fig 3.16: Relative Composition of Total SOC by SOC in Particle Size Fractions in the Prairie Meadows. Relative amounts of SOC represent g C kg-1 soil in each fraction.

117 Relative Compositon of Total SOC by Particle Size Fractions (%) 0% 20% 40% 60% 80% 100%

10

20

40 Clay Silt Fine Sand 60 Coarse Sand Depth (cm)

80

100

Fig 3.17: Relative Composition of Total SOC by SOC in Particle Size Fractions in the Wheat Fields. Relative amounts of SOC represent g C kg-1 soil in each fraction.

118 3.5 Conclusions The data collected in this study confirms the hypotheses tested. There has been a significant loss of SOC in the AG fields, due to the conversion from prairie to annual agriculture. The prairie meadows have also maintained relatively high levels of all of the soil parameters measured, despite the long-term annual harvest of aboveground biomass without fertilizer inputs. The PM sites had 4 times as much MBC and 2 times as much

C in the fine sand POM fraction than the AG sites, in the top 10cm of the soil. The SOC pools in the PM sites were 42% larger and the TSN pools 39% larger than the AG sites, to a depth of 40cm. Particle size fractionation of the SOC revealed losses of 49% of the

C in the coarse sand fraction, 52% of the C in the fine sand fraction, 51% of the C in the silt fraction and 26% of the C in the clay fraction. Clay sized particles were found to be the dominant fraction of the total SOC, with the relative proportion of SOC in the clay fraction increasing in the long-term wheat fields.

The measured SOC losses reflect a number of ecological changes in the transition from the tallgrass prairie to the wheat production system. Large quantities of SOC were likely lost due to the initial disturbance of cultivation on the soil (Davidson and

Ackerman 1993). Reduced C inputs and decreased limitations on belowground decomposition then contributed to further SOC losses in the soils under wheat (Huggins et al. 1998; Buyanovsky et al. 1987). The five study sites in this research contain five different soil series, representing three different soil orders. Even with this inherent landscape variability, the changes in C cycling between the prairie meadows and wheat fields were so extensive, that the pattern of change in SOC was consistent across sites and soil types.

119 This study adds to the body of literature describing losses from SOC pools and changes in SOC cycling in the conversion of natural systems to agriculture. This data also supports the possibility that temperate grasslands can support long-term biomass production. It seems that the large belowground allocations of C in these systems allow them to support the export of aboveground biomass, while maintaining robust belowground SOC pools and ecosystems processes.

References: Allison, V.J., Miller, R.M., Jastrow, J.D., Matamala, R., and Zak, D.R. 2005. Changes in soil microbial community structure in a tallgrass prairie chronosequence. Soil Science Society of America Journal 69: 1412-1421. Bennett, H.H. 1947. Soil conservation in the world ahead. Journal of Soil and Water Conservation 2: 43-50. Blevins, R.L., Lal, R., Doran, J.W., Langdale, G.W., and Frye, W.W. 1998. Conservation tillage for erosion control and soil quality. In Pierce, F.J. and Frye, W.W. (Eds) Advances in Soil and Water Conservation. Chelsea Press, Ann Arbor, MI. Buyanovsky, G.A., Kucera, C.L., and Wagner, G.H. 1987 Comparative analyses of carbon dynamics in native and cultivated ecosytems. Ecology 68: 2023-2031. Cambardella, C. A. and Elliott, E. T. 1992. Particulate soil organic matter changes across a grassland cultivation sequence. Soil Science Society of America Journal 56: 777-783. Cambardella, C. A. and Elliott, E. T. 1993. Carbon and nitrogen distribution in aggregates from cultivated and native grassland soils. Soil Science Society of America Journal 57: 1071-1076. Christensen, B.T. 1992. Physical fractionation of soil and organic matter in primary particle size and density separates. Advances in Soil Science 20:2-90. Cox, T.S., Glover, J.D., Van Tassel, D.L., Cox, C.M., and DeHaan, L.R. 2006. Prospects for Developing Perennial Grain Crops. Bioscience 56: 649-659. Culman, S.W., DuPont, S.T., Glover, J.D., Buckley, D.H., Fick, G.W., Ferris, H., and Crews, T.E. 2009. Long-term impacts of high-input annual cropping and unfertilized perennial grass production on soil properties and belowground food webs in Kansas, USA. Agricullture, Ecosystems, and Environment. Submitted. Davidson, E.A., and Ackerman, I.L. 1993. Changes in soil carbon inventories following the cultivation of previously untilled soils. Biogeochemistry 20: 161-193. Drinkwater, L.E., Wagoner, P., and Sarrantonio, M. 1998. Legume-based cropping systems have reduced carbon and nitrogen losses. Nature 396: 262-65. Fargione, J., Hill, J., Tilman, D., Polasky, S., and Hawthorne, P. 2006. Land clearing 120 and the biofuel carbon debt. Science 319: 1235-1237. Gantzer, C.J., Anderson, S.H., Thompson, A.L. and Brown, J.R. 1990. Estimating soil erosion after 100 years of cropping on Sanborn Field. Journal of Soil and Water Conservation 45: 641-644. Gantzer, C.J., Anderson, S.H., Thompson, A.L., and Brown, J.R. 1991. Evaluation of soil loss after 100 years of soil and crop management. Agronomy Journal 83: 74- 77. Garcia, F.O. and Rice, C.W. 1994. Microbial biomass dynamics in tallgrass prairie. Soil Science Society of America Journal 58: 816-823. Glover, J.D., Cox, C.M., and Reganold, J.P. 2007. Future farming: A return to roots? Scientific American 297: 66-73. Glover, J.D., Culman, S.W., DuPont, S.T., Broussard, W., Young, L., Mangan, M.E., Mai, J.G., Crews, T.E., DeHaan, L.R., Buckley, D.H., Ferris, H., Turner, R.E., Reynolds, H.L., Wyse, D.L., 2009. Harvested perennial grasslands provide ecological benchmarks for agricultural sustainability (submitted to Agriculture, Ecosystem, and Environment). Grandy, A.S. and Robertosn, G.P. 2007. Land-use intensity effects on soil organic carbon accumulation rates and mechanisms. Guo, L.B. and Gifford, R.M. 2002. Soil carbon stocks and land use change: a meta analysis. Global Change Biology 8: 345-360. Gupta, V. V. S. R. and Germida, J. J. 1988. Distribution of microbial biomass and its activity in different soil aggregate size classes as affected by cultivation. Soil Biology and Biochemistry 20: 777-786. Habekost, M., Eisenhauer, N., Scheu, S., Steinbeiss, S, Weigelt, A., and Gleixner, G. 2008. Seasonal changes in the soil microbial community in a grassland plant diversity gradient four years after establishment. Soil Biology and Biochemistry 40: 2588-2595. Howard, A. 1940. An agricultural testament. Oxford University Press, London, U.K. Huggins, D.R., Buyanovsky, G.A., Wagner, G.H., Brown, J.R., Darmody, R.G., Peck, T.R., Lesoing, G.W., Vanotti, M.B., and Bundy, L.G. 1998. Soil organic C in the tallgrass prairie-derived region of the corn belt: effects of long term crop management. Soil and Tillage Research 47: 219-234. IPCC. 2007. Climate Change 2007: Synthesis Report. Fourth assessment report of the Intergovernmental Panel on Climate Change. Jackson, W. 2002. Natural systems agriculture: a truly radical alternative. Agriculture Ecosystems, and Environment 88: 111-117. Lal, R. 2004. Soil carbon sequestration impacts on global climate change and food security. Science 304: 1623-1627. Lal, R. 2007a. Soils and sustainable agriculture: A review. Agronomy for Sustainable Development 27: 1-8. Lal, R. 2007b. Soil Science and the Carbon Civilization. Soil Science Society of America Journal 71: 1425-1437. Lal, R., Kimble, J.M., Follett, R.F., and Cole, C.V. 1999. The Potential of U.S. Cropland to Sequester Carbon and Mitigate the Greenhouse Effect. Boca Raton, FL. CRC Press. 121 Lutzow, M. V., Kogel-Knabner, I., Ekschmitt, K., Flessa, H., Guggenberger, G., Matzner, E., and Marschner. B. 2007. SOM fractionation methods: Relevance to functional pools and to stabilization mechanisms. Soil Biology and Biochemistry. 39:2183-2207. McLauchlan. K. 2006. The nature and longevity of agricultural impacts on soil carbon and nutrients: A review. Ecosystems 9: 1364-1382. Montgomery, D.R. 2007. Dirt: The erosion of civilizations. University of California Press, Berkeley, CA. Moore, J.M., Klose, S., and Tabatabai, M.A. 2000. Soil microbial biomass carbon and nitrogen as affected by cropping systems. Biology and Fertility of Soils 31: 200- 210. Potthoff, M., Steenwerth, K.L, Jackson, L.E., Drenovsky, R.E., Scow, K.M., and Joergensen, R.G. 2006. Soil microbial community composition as affected by restoration practices in California grassland. Soil Biology and Biochemistry 38: 1851-1860. Reganold, J.P., Elliott, L.F., and Unger, Y.L. 1987. Long-term effects of organic and conventional farming on soil erosion. Nature 330: 370-72. Rutledge, E.M., Wilding, L.P., and Elfield, M. 1967. Automated particle size separation by sedimentation. Soil Science Society of America Proceedings 31: 287-288. SAS Institute, 2002. The SAS System for Microsoft Windows Release 8.2 SAS Institute, Cary, NC. Schmer, M.R., Vogel, K.P., Mitchell, R.B., and Perrin, R.K., 2008. Net energy of cellulosic ethanol from switchgrass. Proceedings of the National Academy of Sciences U.S.A. 105: 464-469. Smith, J.S., and Paul, E.A., 1990. The significance of soil microbial biomass estimations. In Bollage, J.M., and Stotzky, G. (Eds) Soil Biochemistry. Marcel Dekker, Inc., New York, NY. Sollins, P., Glassman, C., Paul, E.A., Swanston, C., Ljtha, K., Heil, J.W., and Elliot, E.T. 1999. Soil carbon and nitrogen fractions. In: Robertson, G.P., Coleman, D.C., Bledsoe, C.S., Sollins, P. (Eds.), Standard Soil Methods for Long-term Ecological Research. Oxford University Press, New York, NY, pp. 89-105. Tiessen, H., and Stewart, J.W.B. 1983. Particle size fractions and their use in studies of soil organic matter: II. Cultivation effects on organic matter composition in size fractions. Soil Science Society of America Journal 47: 509-514. Tiessen, H., Cuevas, E., and Chacon, P. 1994. The role of soil organic matter in sustaining soil fertility. Nature 317: 783-785. Tilman, D., Hill, J. and Lehman, C. 2006. Carbon-negative biofuels from low-input high diversity grassland biomass. Science 314: 1598-1600. Vance, E.D., P.C. Brookes, and Jenkinson, D.S. 1987. An extraction method for measuring soil microbial biomass C. Soil Biology and Biochemistry19:703-707. Voroney, R.P., Winter, J.P., and Beyaert, R.P. 1993. Soil microbial biomass C and N. p. 277-286. In M.R. Carter (ed.) Soil sampling and methods of analysis. Lewis Publishers, Boca Raton, FL. Wander, M.M., and Bollero, G.A. 1999. Soil Quality Assessment of Tillage Impacts in Illinois. Soil Science Society of America Journal 63: 961-971. 122 Wardle, D.A., Yeates, G.W., Watson, R.N., and Nicholson, K.S. 1995. The detritus food-web and the diversity of soil fauna as indicators of disturbance regimes in agroecosystems. Plant and Soil 170: 35-43. Warembourg, F.R. and Esterlich, H.D. 2001. Plant phenology and soil fertility effects on below-ground carbon allocation for an annual (Bromus madritensis) and a perennial (Bromus erectus) grass species. Soil Biology and Biochemistry 33: 1291-1303. Wilson, T.C., Paul, E.A., and Harwood, R.R. 2001. Biologically active soil organic matter fractions in sustainable cropping systems. Applied Soil Ecology 16: 63- 76.

123 CHAPTER 4

THE SHORT TERM EFFECTS OF THE CONVERSION FROM TALLGRASS

PRAIRIE TO WHEAT PRODUCTION ON SOIL ORGANIC CARBON

IN NORTH CENRAL KANSAS

4.1 Abstract

This study describes and quantifies the short-term effects of the conversion of an annually harvested tallgrass prairie remnant to annual wheat (Triticum aestivum) production on SOC pools and fractions. A no-till approach to the land conversion was applied in an attempt to control the SOC disturbance caused by tillage. A replicated complete block (n = 3) experiment was established and plots in the prairie meadow (PM) were converted to wheat production (NT) through herbicide application in the summer of

2004.

Soil core samples were collected to a depth of 1m in May and June of 2008.

Management effects on SOC pools were assessed by analyzing total soil organic C

(SOC), total soil nitrogen (TSN), microbial biomass C (MBC) and a particle size fractionation of SOC in coarse sand (>250µm), fine sand (250-53µm), silt (53-2µm), and clay (<2µm) sized fractions. Total SOC, TSN, and all particle size fractions did not show any significant decreases, 3 growing seasons after the no-till conversion. MBC showed a 124 significant decrease in the NT plots to a depth of 40cm. The decrease in microbial activity may be due to large losses of root biomass and the application of agronomic management. These results demonstrated that a no-till conversion of landscapes reduces the impact on SOC pools, but more labile pools such as the MBC were heavily affected and may be predictive of future degradation in the system.

4.2 Introduction The conversion of native ecosystems, composed of perennial plant communities, to production agricultural systems of annual plants is known to cause great reductions in the antecedent SOC pools (Guo and Gifford 2002, Davidson and Ackerman 1993). In temperate regions approximately 20-40% of SOC can be expected to be lost within 5-20 years of the onset of agriculture (Davidson and Ackerman 1993). Large losses of SOC are driven in these systems by disturbances associated with the tillage of the soil

(Reicosky et al. 1997). A reduction C inputs and increased decomposition in the transition from native ecosystems to agricultural systems are additional factors leading to

SOC losses (Huggins et al. 1998). These altered ecosystem processes can degrade soil quality and ecosystem services. Reductions in ecosystem level functioning have been observed in the loss of macro-aggregate structure and microbial biomass (Gupta and

Germida 1988), heavy losses of particulate organic matter (Cambardella and Elliot 1992), and reductions in the complexity and stability of soil food webs (Culman et al. 2009) in studies examining the long-term effects of the cultivation of native ecosystems on the soil system.

125 Conservation agricultural systems in which tillage is reduced, crop residues are retained, and crop rotations are used have shown promise in reducing SOC losses associated with agriculture (Paustian et al. 1997). Eliminating tillage, specifically, from management regimes has demonstrated significant positive impacts on soil quality in soils under no-till management (Blevins et al. 1998, Wander and Bollero 1999). No-till management has also been promoted as a method for increasing SOC stocks in agricultural soils (Lal 2004). The observed effects of no-till management on SOC stocks in the literature have varied widely from studies observing dramatic increases of SOC under no-till (Sa et al 2001), to studies seeing no difference in SOC in different tillage regimes (Blanco-Canqui and Lal 2008) and studies that observed comparatively higher increases in SOC under fields receiving tillage (Gregorich et al. 2009). These discrepancies suggest that the effect of tillage systems on SOC stock is also dependent on other ecological factors including soil type, soil texture and climate. Despite the variable effect of no-till management on SOC stock, the marked long-term increases in soil quality under no-till management make it a valuable agronomic strategy.

While the effect of transitioning previously cultivated agricultural soils to no-till management has been widely studied, very little literature on the effect of converting perennial vegetation to no-till annual agriculture, in a temperate climate, exists.

Franzluebbers and Stuedemann (2008) conducted a study in which 20-year old tall fescue

(Lolium arundinaceum) pastures were converted to annual grain agriculture through both moldboard plowing and an herbicide based no-till approach. Changes in total SOC,

POM-C and MBC were monitored in the agricultural soils continuously for 3 years

(Franzluebbers and Stuedemann 2008). After 3 years, no-till plots showed slight 126 decreases in MBC and in POM-C, and no decrease in total SOC compared with pasture, while the plowed plots showed significant decreases in MBC, POM-C, and total SOC compared with the no-till (Franzluebbers and Stuedemann 2008).

Changes in SOC fractions in annually harvested tallgrass prairie meadows, in north central Kansas, converted directly to no-till wheat (Triticum aestivum) wheat production in 2003 were quantified in this study. Glover et al. (2009) documented the superior ecosystem services provided by long-term annually harvested prairie meadows, compared with fields that have been in annual wheat production for a similar length of time. This no-till conversion study was initiated to attempt to control the ecological degradation associated with soil cultivation in the long-term wheat fields, so that a clearer view of the ecological changes associated with replacing a diverse perennial plant community with a monoculture of annual grains might be obtained. Early research on the conversion site found a 43% reduction in root biomass and significant changes in the soil food web in the no-till plots, compared with the prairie meadows (DuPont et al.

2009).

This study analyzed soils from the conversion experiment with the objective of determining the effects of the direct conversion of prairie meadows to no-till wheat production on active (MBC), intermediate (POM-C), and more stable (mineral associated

SOC, and total SOC) SOC fractions. The hypothesis tested is that the no-till conversion will cause reductions in MBC and POM-C, while not affecting total SOC or mineral associated SOC.

127 4.3 Methods 4.3.1 Study Site The conversion study is located in an uncultivated remnant tallgrass prairie meadow in Niles, Ottawa county, Kansas (N 38° 58’145”, W 97°28’616”). The meadow has been harvested for hay, on an annual basis for over 75 years. The vegetation is mowed to a height of 8-10cm each summer and removed for livestock fodder. The landowners have indicated that fertilizer has never been applied at the site, but it has been burned periodically.

The tallgrass prairie plant community at the site has the following ground cover based composition: warm season native grasses 73%, native perennial legumes 9%, native non-leguminous perennial forbs 2%, and non-native annual grasses 11%. The site sits on a fertile bottomland soil designated as Class 1 prime farmland by the USDA

Natural Resource Conservation Service. The soil series is Hord; a Fine-silty, mixed, superactive, mesic Cumulic Haplustoll, formed on an alluvial terrace. Mean annual temperature in the region is 13.2°C and the mean annual precipitation is 730mm.

In Summer 2003 three replicate blocks were established in a randomized complete block design. Each block contained two 10m by 20m plots, with randomly assigned treatments. One plot in each block was assigned the annual crop, no-till (NT) treatment, while the other plot remained in the harvested prairie meadow (PM) treatment.

The vegetation in the NT plots were sprayed multiple times with herbicide, before no-till planting began with soybeans (Glycine max) in 2004. Soybeans were planted in the NT plots again in 2005, with sorghum (Sorghum bicolor) and wheat (Triticum aestivum)

128 following in 2006, wheat in 2007 and wheat in 2008. Table 2.1 summarizes the management of the plots.

4.3.2 Soil Sampling Soil sampling was carried out in mid May (peak wheat biomass) and late June (peak prairie biomass) 2008. Three 4cm diameter soil cores were taken in each treatment field to a depth of 1m, along a transect, in each plot. Cores were divided into 0-10, 10-20, 20-

40, 40-60, 60-80, and 80-100cm depths. The three cores for each plot were then bulked according to depth and mixed until homogenous. The mass of cores was recorded for

each depth for bulk density (ρb) measurements. 100g sub-samples were taken from each sample and dried for 24h at 105°C to determine the gravimetric moisture content (ω).

129

Year Season Operation Product Rate or yield (kg ha-1)

2003 Summer Herbicide (3X) Glyphosate 0.25 a.i.*

2004 Spring Herbicide (3X) Glyphosate 0.25 a.i.*

Summer Plant Soybeans Crop failed due to

weed pressure.

2005 Spring Herbicide (3X) Glyphosate 0.25 a.i.*

Summer Plant Soybeans 65

Fall Harvest Soybeans 584

2006 Spring Herbicide Glyphosate 0.25 a.i.*

Summer Plant Sorghum cover crop 8

Fall Plant Winter Wheat 135

Fall Fertilize Monoammonium 65

Phosphate

2007 Spring Fertilize Urea 112

Summer Harvest Wheat 1747

*a.i. = active ingredient of herbicide products.

Table 4.1 Management of no-till conversion plots 2003-2007 (from DuPont et al.

2009). 130 4.3.3 Total C and Total N Air-dried soil, from the June sampling date, was passed through a 2mm sieve and ground with a mortar and pestle in preparation for C/N analysis. Samples in the range of

40-60mg, based on estimated C and N content, were weighed and encased in 5x9mm

Costech sampling. These samples were then shipped to the Stable Isotope Facility at the

University of California at Davis, where they were analyzed for total soil C and total soil

N (TSN) by combustion, using a PDZ Europa ANCA-GSL elemental analyzer (Sercon

Ltd., Chesire, UK). Inorganic carbonate composition in these soils is negligible and total

C is considered soil organic C (SOC) in this study. SOC and TSN pools were calculated according to the following equation (Eq 1) (Lal et al 1999):

Mg C or N ha-1 = %C or N × B.D. (Mg m-3) × depth (m) × 104m2 ha-1 100 ………..(Eq 1)

4.3.4 Microbial Biomass C Microbial biomass carbon (MBC) was obtained using the chloroform fumigation extraction method (Vance 1987). Briefly, three 10g samples of field moist soil, sieved through at 6.75mm sieve, were taken from each sample. One 10g sample served as a control, one as a fumigated sample, and the third was used for moisture content.

The fumigated samples were place in a vacuum dessicator with approximately 50ml of chloroform. The dessicator was evacuated until the chloroform boiled for 1 minute.

The dessicator was then sealed and the samples were left in the dessicator for 24 hours.

Following the fumigation, the samples were extracted by placing them in 250ml plastic 131 bottles with 80ml of .05M K2(SO4). The bottles were then shaken in the floor shaker for

1 hour. The shaken samples were centrifuged for 5 minutes and the supernatant was passed through at .45µm syringe filter. Control samples were extracted using the same process, but were not fumigated. Two analytical replicates were analyzed for each sampling date, and blanks were prepared with each replicate.

Filtered samples were frozen in 50ml centrifuge tubes. Frozen samples were sent to the Stable Isotope Facility at the University of California at Davis, where they were analyzed for total organic carbon (TOC) with an O.I. Analytical Model 1010 TOC

Analyzer (OI Analytical, College Station, TX).

MBC was calculated using equations from the method of Voroney et al. (1993). All soil weights were corrected for moisture content. Total weight of extractable C in the fumigated (OF) and unfumigated (OUF) soil samples:

-1 -1 OCF, OCUF (µg g soil ) = extractable C (ug mL ) X VS (ml) / MS (g)

(Eq 2)

Where VS is the volume of the extractant solution and MS is the mass of soil in the solution.

Microbial Biomass C in the soil (MBC):

-1 MBC (ug g soil ) = (OCF – OCUF)…………………………………(Eq 3)

132 4.3.5 Particle Size Fractionation A primary particle size fractionation (Christensen 1992) was carried out to divide the soil into coarse sand (2mm-250µm), fine sand (250-53µm), silt (53-2µm), and clay

(<2µm) sized fractions. The soil was dispersed chemically, with Na-HMP, using a method outlined in Cambardella and Elliot (1992) and Sollins et al (1999). Air-dried soil, from the June sampling date, was passed through a 2mm sieve. 20g of soil were placed in a 250ml plastic bottle in 60ml of 5g L-1 Na-HMP and shaken overnight (16h) in a floor shaker. The resulting solution was then passed through nested 250µm and 53µm sieves.

DI water was used to wash the solution through the sieves and the material captured on the sieves was washed into pre-weighed glass beakers and oven dried (<45°C). The resulting material was considered as the sand sized fractions and POM. The soil slurry that passed through the sieves was collected in plastic cylinders and separated into silt and clay by sequential decantation and siphoning (Rutledge et al 1967). The silt remaining in the cylinders was washed into pre-weighed glass beakers and oven dried.

The clay suspension was flocculated using 0.5M MgCl2 prior to oven drying. The coarse sand (CS), fine sand (FS), and silt (SI) were weighed to calculate their relative composition to the whole soil, and the mass of the clay (CL) fraction was obtained by subtraction.

Fractionated samples were ground with a mortar and pestle, and 40-60mg samples were prepared, based on estimated C and N content, for C/N analysis. Samples were packed in Costech 5×9mm sample tins and shipped to the Stable Isotope Facility at the

University of California at Davis where they were analyzed for total C and total N by

133 combustion, using a PDZ Europa ANCA-GSL elemental analyzer (Sercon Ltd., Chesire,

UK).

C and N content in each fraction were calculated (in g C or N kg soil-1) using the following equation from Sollins et al (1999):

C(fr) = Cfr x Wfr x 10……………….………………………………..(Eq 4)

Where the C, or N, in each fraction, C(fr), is equal to the % concentration of C in that

fraction (Cfr) multiplied by the dry mass of that fraction (Wfr) (g fraction/g soil) multiplied by ten to achieve C in g kg soil-1.

4.3.6 Data Analysis All data was analyzed using analysis of variance (ANOVA) using the PROC MIXED procedure in SAS software (Cary, N.C.). Depth and treatment were treated as fixed effects, and site was treated as a random effect, in the analysis. Statistical significance was determined at α = 0.05 level.

4.4 Results 4.4.1 Soil Organic Carbon Total SOC in the conversion study plots showed no detectable changes in either concentration (Fig 4.1) or pools (Table 4.2), three growing seasons after the conversion to no-till agriculture. Total SOC changes relatively slowly under most management changes, so these results are not surprising. It appears that without the large disturbance of the soil surface that occurs with tillage, there is not a large release of SOC during the initiation of no-till agriculture.

134 SOC (%) 0 0.5 1 1.5 2 2.5 3

10

20

40 Prairie Meadow No-till 60 Depth (cm)

80

100

Figure 4.1: Depth Profile of Soil Organic C Concentration (%) in the conversion study plots. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Bars indicate the standard error.

135

SOC Pools (Mg C ha-1)

Depth (cm) Prairie No-Till S.E.

0-10 31.56 31.32 ±2.59

10-20 30.63 30.20 ±2.22

20-40 28.76 27.94 ±1.68

40-60 26.89 25.68 ±1.63

60-80 25.02 23.43 ±2.11

80-100 23.15 21.17 ±2.86

Table 4.2: Depth Distribution of SOC pools (Mg C ha-1) in the conversion study plots. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1.

136 SOC Pools (Mg/ha) 0 10 20 30 40

10

20

40 Prairie Meadows Wheat Fields 60 Depth (cm)

80

100

Figure 4.2: Depth Profile of SOC Pools (Mg C ha-1) in the conversion study plots. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Bars indicate the standard error.

137 4.4.2 Total Soil Nitrogen No changes in TSN were detectable in either concentration (Fig 4.3) or TSN pools

(Table 4.3). The values recorded for the TSN pools are almost identical for the two treatments (Table 4.3).

TSN (%) 0 0.05 0.1 0.15 0.2 0.25 0.3

10

20

40 Prairie Meadows Wheat Fields 60 Depth (cm)

80

100

Figure 4.3: Depth Profile of TSN Concentration (%) in the conversion study plots. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Bars indicate the standard error.

138

Total Soil N Pools (Mg N ha-1)

Depth (cm) Prairie No-Till S.E.

0-10 2.97 2.94 ±0,23

10-20 2.88 2.84 ±0.20

20-40 2.68 2.64 ±0.15

40-60 2.48 2.44 ±0.15

60-80 2.28 2.25 ±0.19

80-100 2.08 2.05 ±0.26

Table 4.3: Depth Distribution of TSN pools (Mg N ha-1) in the conversion study plots. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1.

139 TSN (Mg N/ha) 0 1 2 3 4

10

20

40 Prairie Meadow No-till 60 Depth (cm)

80

100

Figure 4.4: Depth Profile of TSN pools (Mg N ha-1) in the conversion study plots. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Bars indicate the standard error.

140 4.4.3 Microbial Biomass C Significant decreases in MBC were detected to 40cm depth in the NT treatments (Fig

4.5 and Table 4.4), 3 full growing seasons after the conversion from prairie to no-till agriculture. This result is in line with Franzluebbers and Stuedemann (2008) which also observed changes in MBC, but no change in total SOC, 3 years after converting perennial pasture to grain production in Georgia. Given that total SOC showed no change, the change in MBC may be an early indication of long-term changes that will occur in the

SOC at this site (Powlson et al. 1987). MBC plays a central role in SOC storage and cycling, nutrient cycling, and supports larger organism in the soil food web (Dalal 1998).

So the reductions in MBC observed here may serve as an early indication of degraded ecosystem processes in the transition from prairie to wheat production.

Agronomic management practices, such as the application of herbicide and inorganic

N fertilizer that occurred in the conversion from prairie to wheat production, may have been factors in the loss of MBC in the no-till plots. A number of previous studies have indicated that MBC levels are not reduced by the application of glyphosate based herbicides, which were used in this experiment (Wardle and Parkinson 1990; Haney et al.

2000; and Laupwayi et al. 2009). This suggests that factors other than herbicide application contributed to the observed loss of MBC. A significant number of studies also exist where the application of inorganic N fertilizer application has led to observations of reduced MBC (Garcia and Rice 1994; Omay et al. 1997; and Maly et al.

2009). This may be due to increased mineralization of available C substrates, with increased N (Maly et al. 2009).

141 The conversion of the tallgrass prairie represents a major change in numerous ecosystem properties and processes that may have affected MBC levels. MBC levels are known to be associated with the availability of organic compounds from belowground detritus, derived form plant litter and roots (Smith and Paul 1990, Dalal et al. 1995).

Allocation of C to detritus pools is known to be higher under tallgrass prairie than in wheat systems (Buyanovsky et al. 1987) and studies have observed higher levels of MBC in temperate grasslands compared with cropping systems in similar soils (Smith and Paul

1990, Dalal et al. 1995). Dalal et al. (1995) indicated strong correlations between MBC and root biomass. This may be a significant factor in MBC levels at this site, where root biomass was observed to be reduced by 43% in the no-till plots (Dupont et al. 2009). A study of the relationships between plant diversity and soil microbial communities in experimental grasslands in Minnesota indicated reduced microbial biomass associated with reductions in the diversity of plant communities (Zak et al. 2003). This result was correlated with increased primary productivity in the more diverse plant communities

(Zak et al. 2003). Given the high level of complexity inherent in the conversion ecosystems, it is likely that the observed changes in MBC were driven by a number of factors.

142 MBC (ug/g) 0 50 100 150 200 250 300

* 10

* 20

Prairie Meadows 40 * Wheat Fields Depth (cm) 60

80

Figure 4.5: Depth Profile of Microbial Biomass C (µg C g soil-1) in conversion study plots. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Asterisk indicates mean values are significantly different (p <.05).

143

Microbial Biomass C (µg C g-1 soil)

Depth (cm) Prairie No-Till S.E.

0-10 *266.46 207.85 ±15.92

10-20 *235.02 184.59 ±13.64

20-40 *172.14 138.05 ±10.30

40-60 109.27 91.59 ±10.00

60-80 46.39 44.97 ±12.94

Table 4.4: Depth Distribution of MBC (µg C g soil-1) in conversion study plots. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1. Asterisk indicates mean values are significantly different (p <.05).

144 4.4.4 Particle Size Fractionation 4.4.4.1 The Coarse Sand Fraction No significant differences were detected in SOC in the coarse sand (>250µm) fraction.

These larger, sand-sized particles, have a relatively short turnover time (Lutzow et al

2007). It is somewhat unexpected that no differences were observed despite the high level of disturbance that occurred in the system with the conversion from prairie to wheat.

SOC in the Coarse Sand Fraction (g C / kg soil) 0 0.5 1 1.5

10

20

Prairie Meadows 40 No Till Depth (cm) 60

80

Figure 4.6: Depth Profile SOC in the Coarse Sand (>250µm) Fraction (g C kg-1 soil) measured in conversion study plots. Values are means from ANOVA run in PROC MIXED, SAS v9.1. Bars indicate the standard error.

145 TSN in the Coarse Sand Fraction (g N / kg soil) 0 0.01 0.02 0.03 0.04 0.05 0.06

10

20

Prairie Meadows 40 No Till Depth (cm) 60

80

Figure 4.7: Depth Profile TSN in the Coarse Sand (>250µm) Fraction (g N kg-1 soil) measured in conversion study plots. Values are means from ANOVA run in PROC MIXED, SAS v9.1. Bars indicate the standard error.

146 4.4.4.2 The Fine Sand Fraction No significant differences were detected in the results for both C and N in the POM in the fine sand fraction (Fig 4.8 and Fig. 4.9). Franzluebbers and Stuedemann (2008) did not observe any significant differences in POM-C in their no-till conversion study. These results suggest that the POM is not heavily affected by non-tillage land conversions. This fraction has show large immediate reductions in Midwestern grasslands converted to crop production using tillage (Cambardella and Elliot 1992). The disruption of the soil aggregate structure associated with tillage may lead to increased availability and mineralization of POM. These ecological disturbances of the soil system may not occur, or may occur on a longer timeframe, with a no-till conversion.

C (g C/kg soil) 0 0.5 1 1.5 2 2.5 3 3.5

10

20

Prairie Meadow 40 No-till Depth (cm) 60

80

Figure 4.8: Depth Profile SOC in the Fine Sand (250-53µm) Fraction (g C kg-1 soil) measured in conversion study plots. Values are means from ANOVA run in PROC MIXED, SAS v9.1. Bars indicate the standard error.

147

SOC in the Fine Sand (250-53µm) Fraction (g C kg-1 soil)

Depth (cm) Prairie No-Till S.E.

0-10 2.24 2.75 ±0.43

10-20 1.96 2.38 ±0.37

20-40 1.40 1.64 ±0.28

40-60 0.84 0.90 ±0.35

60-80 0.28 0.16 ±0.47

Table 4.5: Depth Distribution of SOC in the Fine Sand (250-53µm) Fraction (g C kg- 1 soil) in conversion study plots. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1.

148 N (g N/kg soil) 0 0.05 0.1 0.15 0.2 0.25

10

20

Prairie Meadow 40 No-till Depth (cm) 60

80

Figure 4.9: Depth Profile TSN in the Fine Sand (250-53µm) Fraction (g N kg-1 soil) measured in conversion study plots. Values are means from ANOVA run in PROC MIXED, SAS v9.1. Bars indicate the standard error.

149 4.4.4.3 The Silt Fraction No changes in C or N concentration (Figs 4.10 and Fig 4.11) were detected in the silt fraction (53-2µm). This fraction is thought to have a relatively slow turnover, and long residence time (Christensen 1992, Lutzow et al. 2007). It appears, that after 3 years, the disturbance of the no-till conversion has had little effect on this SOC fraction.

SOC in the Silt (53-2µm) Fraction (g C kg-1 soil)

Depth (cm) Prairie No-Till S.E.

0-10 8.85 9.10 ±0.73

10-20 7.87 8.08 ±0.62

20-40 5.91 6.05 ±0.47

40-60 3.94 4.02 ±0.46

60-80 1.97 1.99 ±0.59

80-100 0.05 0.05 ±0.80

Table 4.6: Depth Distribution of SOC in the Silt (53-2µm) Fraction (g C kg-1 soil) in conversion study plots. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1.

150 SOC in the Silt Fraction (g C / kg soil) 0 2 4 6 8 10 12

10

20

40 Prairie Meadows No Till 60 Depth (cm)

80

100

Figure 4.10: Depth Profile of SOC in the Silt (53-2µm) Fraction (g C kg-1 soil) in conversion study plots. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1.

151 TSN in the Silt Fraction (g N / kg soil) 0 0.2 0.4 0.6 0.8 1

10

20

40 Prairie Meadows No Till 60 Depth (cm)

80

100

Figure 4.11: Depth Profile of TSN in the Silt (53-2µm) Fraction (g C kg-1 soil) in conversion study plots. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1.

152 4.4.4.4 The Clay Fraction No significant differences were found in the C or N concentrations in the clay fraction

(Figs 4.12 and 4.13). The clay associated SOC is also considered to have a slow turnover, and long residence time (Christensen 1992, Lutzow et al. 2007). This result indicates that the no-till conversion of the prairie meadow to wheat production had little or no effect on this fraction.

SOC in the Clay (<2µm) Fraction (g C kg-1 soil)

Depth (cm) Prairie No-Till S.E.

0-10 13.08 12.66 ±0.47

10-20 12.10 11.72 ±0.40

20-40 10.14 9.84 ±0.30

40-60 8.17 7.96 ±0.30

60-80 6.21 6.09 ±0.38

80-100 4.25 4.21 ±0.52

Table 4.7: Depth Distribution of SOC in the Clay (<2µm) Fraction (g C kg-1 soil) in conversion study plots. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1.

153 SOC in the Clay Fraction (g C / kg soil) 0 5 10 15

10

20

40 Prairie Meadows No Till 60 Depth (cm)

80

100

Figure 4.12: Depth Profile of SOC in the Clay (<2µm) Fraction (g C kg-1 soil) in conversion study plots. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1.

154 TSN in Clay Fraction (g N / kg soil) 0 0.5 1 1.5

10

20

40 Prairie Meadows No Till 60 Depth (cm)

80

100

Figure 4.13: Depth Profile of TSN in the Clay (<2µm) Fraction (g C kg-1 soil) in conversion study plots. Values are means from ANOVA, run in PROC MIXED, SAS v.9.1.

155 4.5 Conclusions In this study of the no-till conversion of harvested prairie meadows to wheat production, many of the soil properties analyzed, including SOC concentration and pools, TSN, and particle size fractions of SOC, did not show significant change. MBC was the only soil property to show significant decreases after the conversion to agriculture. This may be due to MBC’s strong relationship to the root biomass and rhizosphere of the prairie plant community. This result may also support evidence of MBC as an early indicator of long term, management induced SOC changes. The lack of change in total SOC, TSN, and particle size fraction pools demonstrate evidence that a no-till approach may reduce the impact of land conversion on global C and biogeochemical cycles.

References Blanco-Canqui, H., and Lal, R. 2008. No-tillage and soil profile carbon sequestration: An on-farm assessment. Soil Science Sociey of America Journal 72: 693-701. Blevins, R.L., Lal, R., Doran, J.W., Langdale, G.W., and Frye, W.W. 1998. Conservation tillage for erosion control and soil quality. In Pierce, F.J. and Frye, W.W. (Eds) Advances in Soil and Water Conservation. Chelsea Press, Ann Arbor, MI. Buyanovsky, G.A., Kucera, C.L., and Wagner, G.H. 1987 Comparative analyses of carbon dynamics in native and cultivated ecosytems. Ecology 68: 2023-2031. Cambardella, C. A. and Elliott, E. T. 1992. Particulate soil organic matter changes across a grassland cultivation sequence. Soil Science Society of America Journal 56: 777-783. Christensen, B.T. 1992. Physical fractionation of soil organic matter in primary particle size and density separates. Advances in Soil Science 20: 1-90. Culman, S.W., DuPont, S.T., Glover, J.D., Buckley, D.H., Fick, G.W., Ferris, H., Crews, T.E., 2009. Long-term impacts of high-input annual cropping and unfertilized perennial grass production on soil properties and belowground food webs in Kansas, USA. (submitted to Agriculture, Ecosystem, and Environment). Dalal, R.C. 1998. Soil microbial biomass- what do the numbers really mean? Australian Journal of Experimental Agriculture 38: 649-665. Dalal, R.C., Strong, W.M., Weston, E.J., Cahill, M.J., Cooper, J.E., Lehane, K.J., King, A.J. , and Chicken, C.J. 1995. Sustaining productivity of a vertisol at Warra, Queensland, with fertilizers, no-tillage, or legumes. 1. Organic matter status. 156 Australian Journal of Experimental Agriculture 35: 903-913. Davidson, E.A., and Ackerman, I.L. 1993. Changes in soil carbon inventories following the cultivation of previously untilled soils. Biogeochemistry 20: 161-193. DuPont, S.T., Culman, S.W., Ferris, H., Buckley, D.H., Glover, J.D., 2009. No-tillage conversion of hayed perennial grassland to annual cropland reduces root biomass, decreases active carbon stocks, and impacts soil biota. (submitted to Agriculture, Ecosystem, and Environment). Franzluebbers, A.J., and Stuedemann, J.A. 2008. Early responses of soil organic fractions to tillage and integrated crop-livestock production. Soil Science Society of America Journal 72: 613-625. Garcia, F.O. and Rice, C.W. 1994. Microbial biomass dynamics in tallgrass prairie. Soil Science Society of America Journal 58: 816-823. Glover, J.D., Culman, S.W., DuPont, S.T., Broussard, W., Young, L., Mangan, M.E., Mai, J.G., Crews, T.E., DeHaan, L.R., Buckley, D.H., Ferris, H., Turner, R.E., Reynolds, H.L., Wyse, D.L., 2009. Harvested perennial grasslands provide ecological benchmarks for agricultural sustainability (submitted to Agriculture, Ecosystem, and Environment). Gregorich, E.G., Carter, M.R., Angers, D.A., and Drury, C.F. 2009. Using a sequential density and particle size fractionation to evaluate carbon and nitrogen storage in the profile of tilled and no-till soils in eastern Canada. Canadian Journal of Soil Science 89: 255-267. Guo, L.B. and Gifford, R.M. 2002. Soil carbon stocks and land use change: a meta analysis. Global Change Biology 8: 345-360. Gupta, V. V. S. R. and Germida, J. J. 1988. Distribution of microbial biomass and its activity in different soil aggregate size classes as affected by cultivation. Soil Biology and Biochemistry 20: 777-786. Haney, R.L., Senseman, S.A., Hons, F.M., and Zuberer, D.A. 2000. Effect of Glyphosate on soil microbial activity and biomass. Weed Science 48: 89-93. Huggins, D.R., Buyanovsky, G.A., Wagner, G.H., Brown, J.R., Darmody, R.G., Peck, T.R., Lesoing, G.W., Vanotti, M.B., and Bundy, L.G. 1998. Soil organic C in the tallgrass prairie-derived region of the corn belt: effects of long term crop management. Soil and Tillage Research 47: 219-234. Lal, R. 2004. Soil carbon sequestration impacts on global climate change and food security. Science 304: 1623-1627. Lal, R., Kimble, J.M., Follett, R.F., and Cole, C.V. 1999. The Potential of U.S. Cropland to Sequester Carbon and Mitigate the Greenhouse Effect. Boca Raton, FL. CRC Press. Laupwayi, N.Z., Harker, K.N., Clayton, G.W., O’Donovan, J.T., and Blackshaw, R.E. 2009. Soil microbial response to herbicide applied to glyphosate-resistant canola. Agriculture, Ecosystems, and Environment 129: 171-176. Lutzow, M. V., Kogel-Knabner, I., Ekschmitt, K., Flessa, H., Guggenberger, G., Matzner, E., and Marschner. B. 2007. SOM fractionation methods: Relevance to functional pools and to stabilization mechanisms. Soil Biology and Biochemistry. 39:2183-2207. Maly, S., Kralovec, J., and Hampel, D. 2009. Effects of long-term mineral fertilization 157 on microbial biomass, microbial activity and presence of r- and K-strategists. Biology and Fertility of Soils 45: 753-760. Omay, A.B., Rice, C.W., Maddux, L.D., and Gordon, W.B. 1997. Changes in soil microbial and chemical properties under long-term crop rotation and fertiliztion. Soil Science Society of America Journal 61: 1672-1678. Paustian, K., Andren, O., Janzen, H. H., Lal, R., Smith, P., Tian, G., Tiessen, H., Van Noordwijk, M., and Woomer, P.L. 1997. Agricultural soils as a sink to mitigate

CO2 emiisions. Soil Use and Management 13: 230-244. Powlson, D.S., Brookes, P.C., and Christensen, B.T. 1987. Measurements of soil microbial biomass provides an early indication of changes in total soil organic matter due to straw incorporation. Soil Biology and Biochemistry 19: 159-164. Reicosky, D.C., Dugas, W.A., and Torbert, H.A., 1997. Tillage-induced soil carbon dioxide loss from different cropping systems. Soil and Tillage Research 41: 105- 118. Rutledge, E.M., Wilding, L.P., and Elfield, M. 1967. Automated particle size separation by sedimentation. Soil Science Society of America Proceedings 31: 287-288. Sa, J.C. de M., Cerri, C., Dick, W.A., Lal, R., Filho, S.P.V., Piccolo, M.C., and Feigl, B.E. 2001. Organic matter dynamics and carbon sequestration rates for a tillage chronosequence in a Brazilian Oxisol. Soil Science Society of America Journal 65:1486-1499. SAS Institute, 2002. The SAS System for Microsoft Windows Release 8.2 SAS Institute, Cary, NC. Smith, J.L. and Paul, E.A. 1990. The significance of soil biomass estimations. In Bollog, J.M, and Stotzky, G. Eds., Soil Biochemistry, Vol 6. pp 357-396. Marcel Dekker, New York, NY. Sollins, P., Glassman, C., Paul, E.A., Swanston, C., Ljtha, K., Heil, J.W., and Elliot, E.T. 1999. Soil carbon and nitrogen fractions. In: Robertson, G.P., Coleman, D.C., Bledsoe, C.S., Sollins, P. (Eds.), Standard Soil Methods for Long-term Ecological Research. Oxford University Press, New York, NY, pp. 89-105. Vance, E.D., P.C. Brookes, and Jenkinson, D.S. 1987. An extraction method for measuring soil microbial biomass C. Soil Biology and Biochemistry19:703-707. Voroney, R.P., Winter, J.P., and Beyaert, R.P. 1993. Soil microbial biomass C and N. p. 277-286. In M.R. Carter (ed.) Soil sampling and methods of analysis. Lewis Publishers, Boca Raton, FL. Wander, M.M., and Bollero, G.A. 1999. Soil Quality Assessment of Tillage Impacts in Illinois. Soil Science Society of America Journal 63: 961-971. Wardle, D.A., and D. Parkinson. 1990. Effects of three herbicides on soil microbial biomass and activity. Plant and Soil 122: 21-28. Zak, D.R., Holmes, W.E., White, D.C., Peacock, A.D., and Tilman, D. 2003. Plant diversity, soil microbial communities, and ecosystem function: are there any links? Ecology 84: 2042-2050.

158 COMPLETE REFERENCE LIST

Allison, V.J., Miller, R.M., Jastrow, J.D., Matamala, R., and Zak, D.R. 2005. Changes in soil microbial community structure in a tallgrass prairie chronosequence. Soil Science Society of America Journal 69: 1412-1421. Batjes, N.H. 1996. Total carbon and nitrogen in the soils of the world. European Journal of Soil Science 47: 151-163. Bennett, H.H. 1947. Soil conservation in the world ahead. Journal of Soil and Water Conservation 2: 43-50. Blake, G.R., Hartge, K.H., 1986. Bulk Density. In: Klute, A. (Ed.), Methods of Soil Analysis Part 1- Physical and Mineralogical Methods. Soil Science Society of America, Inc., Madison, WI, USA, pp. 364-367. Blanco-Canqui, H., Gantzer, C.J., Anderson, S.H., Alberts, E.E., and Thompson, A.L. 2004. Grass barrier and vegetative filter strip effectiveness in reducing runoff, sediment, nitrogen and phosphorous loss. Soil Science Society of America Journal 68: 1670-1678. Blanco-Canqui, H., and Lal, R. 2008. No-tillage and soil profile carbon sequestration: An on-farm assessment. Soil Science Sociey of America Journal 72: 693-701. Blevins, R.L., Lal, R., Doran, J.W., Langdale, G.W., and Frye, W.W. 1998. Conservation tillage for erosion control and soil quality. In Pierce, F.J. and Frye, W.W. (Eds) Advances in Soil and Water Conservation. Chelsea Press, Ann Arbor, MI. Brye, K.R. and Kucharik, C.J. 2003. Carbon an nitrogen sequestration in two prairie topochronosequences on contrasting soils in southern Wisconsin. American Midland naturalist 149: 90-103. Buyanovsky, G.A., Kucera, C.L., and Wagner, G.H. 1987. Comparative Analysis of Carbon Dynamics in Native and Cultivated Ecosystems. Ecology 68: 2023-2031. Cambardella, C. A. and Elliott, E. T. 1992. Particulate soil organic matter changes across a grassland cultivation sequence. Soil Science Society of America Journal 56: 777-783. Cambardella, C. A. and Elliott, E. T. 1993. Carbon and nitrogen distribution in aggregates from cultivated and native grassland soils. Soil Science Society of America Journal 57: 1071-1076. Christensen, B.T. 1992. The physical fractionation of soil and organic matter into primary particle size and density separates. Advances in Soil Science 20: 1-90. Christensen, B.T. 2001. Physical fractionation of soil and structural and functional complexity in organic matter turnover. European Journal of Soil Science 52: 345- 353. Cox, T.S., Glover, J.D., Van Tassel, D.L., Cox, C.M., and DeHaan, L.R. 2006. 159 Prospects for Developing Perennial Grain Crops. Bioscience 56: 649-659. Culman, S.W., DuPont, S.T., Glover, J.D., Buckley, D.H., Fick, G.W., Ferris, H., and Crews, T.E. 2009. Long-term impacts of high-input annual cropping and unfertilized perennial grass production on soil properties and belowground food webs in Kansas, USA. Agricullture, Ecosystems, and Environment. Submitted. Dalal, R.C. 1998. Soil microbial biomass- what do the numbers really mean? Australian Journal of Experimental Agriculture 38: 649-665. Dalal, R.C., Strong, W.M., Weston, E.J., Cahill, M.J., Cooper, J.E., Lehane, K.J., King, A.J. , and Chicken, C.J. 1995. Sustaining productivity of a vertisol at Warra, Queensland, with fertilizers, no-tillage, or legumes. 1. Organic matter status. Australian Journal of Experimental Agriculture 35: 903-913. Davidson, E.A., and Ackerman, I.L. 1993. Changes in soil carbon inventories following the cultivation of previously untilled soils. Biogeochemistry 20: 161-193. Doran, J.W. and Parkin, T.B. 1994. Defining and Assessing Soil Quality. In Doran, J.W., Coleman, D.C., Bezdicek, D.F., and Stewart, B.A. (Eds). Defining Soil Quality for a Sustainable Environment. Soil Science Society of America, Inc., Madison, WI. Drinkwater, L.E., Wagoner, P., and Sarrantonio, M. 1998. Legume-based cropping systems have reduced carbon and nitrogen losses. Nature 396: 262-65. DuPont, S.T., Culman, S.W., Ferris, H., Buckley, D.H., Glover, J.D., 2009. No-tillage conversion of hayed perennial grassland to annual cropland reduces root biomass, decreases active carbon stocks, and impacts soil biota. (submitted to Agriculture, Ecosystem, and Environment). Fargione, J., Hill, J., Tilman, D., Polasky, S., and Hawthorne, P. 2006. Land clearing and the biofuel carbon debt. Science 319: 1235-1237. Follett, R.F. 2001. Soil Management Concepts and Carbon Sequestration in Cropland Soils. Soil and Tillage Research. 61: 77-92. Fornara, D.A. and Tilman, D. 2008. Plant functional composition influences rates of soil carbon and nitrogen accumulation. Journal of Ecology 96: 314-322. Franzluebbers, A.J., and Stuedemann, J.A. 2008. Early responses of soil organic fractions to tillage and integrated crop-livestock production. Soil Science Society of America Journal 72: 613-625. Gantzer, C.J., Anderson, S.H., Thompson, A.L. and Brown, J.R. 1990. Estimating soil erosion after 100 years of cropping on Sanborn Field. Journal of Soil and Water Conservation 45: 641-644. Gantzer, C.J., Anderson, S.H., Thompson, A.L., and Brown, J.R. 1991. Evaluation of soil loss after 100 years of soil and crop management. Agronomy Journal 83: 74- 77. Garcia, F.O. and Rice, C.W. 1994. Microbial biomass dynamics in tallgrass prairie. Soil Science Society of America Journal 58: 816-823. Glover, J.D., Reganold, J.P., and Andrews, P.K. 2000. Systematic method for rating soil quality of conventional, organic and integrated apple orchards in Washington state. Agriculture, Ecosystems, and Environment 80: 29-45. Glover, J.D., Cox, C.M., and Reganold, J.P. 2007. Future farming: A return to roots? Scientific American 297: 66-73. 160 Glover, J.D., Culman, S.W., DuPont, S.T., Broussard, W., Young, L., Mangan, M.E., Mai, J.G., Crews, T.E., DeHaan, L.R., Buckley, D.H., Ferris, H., Turner, R.E., Reynolds, H.L., Wyse, D.L., 2009. Harvested perennial grasslands provide ecological benchmarks for agricultural sustainability (submitted to Agriculture, Ecosystem, and Environment). Grandy, A.S. and Robertosn, G.P. 2007. Land-use intensity effects on soil organic carbon accumulation rates and mechanisms. Ecosystem 10: 58-73. Gregorich, E.G., Carter, M.R., Angers, D.A., and Drury, C.F. 2009. Using a sequential density and particle size fractionation to evaluate carbon and nitrogen storage in the profile of tilled and no-till soils in eastern Canada. Canadian Journal of Soil Science 89: 255-267. Guo, L.B. and Gifford, R.M. 2002. Soil carbon stocks and land use change: a meta- analysis. Global Change Biology 8: 345-360. Gupta, V. V. S. R. and Germida, J. J. 1988. Distribution of microbial biomass and its activity in different soil aggregate size classes as affected by cultivation. Soil Biology and Biochemistry 20: 777-786. Habekost, M., Eisenhauer, N., Scheu, S., Steinbeiss, S, Weigelt, A., and Gleixner, G. 2008. Seasonal changes in the soil microbial community in a grassland plant diversity gradient four years after establishment. Soil Biology and Biochemistry 40: 2588-2595. Haney, R.L., Senseman, S.A., Hons, F.M., and Zuberer, D.A. 2000. Effect of Glyphosate on soil microbial activity and biomass. Weed Science 48: 89-93. Hassink, J. 1997. The capacity of soils to preserve organic C and N by their association with clay and silt particles. Plant and Soil 191: 77-87. Houghton, R.A. 1995. Changes in storage of terrestrial carbon since 1850. in R. Lal, J. Kimble, E. Levine, and B.A. Stewart (eds) “Soils and Global Change”. Boca Raton, FL: CRC Publishers. Howard, A. 1940. An agricultural testament. Oxford University Press, London, U.K. Hudson, B. D. 1994. Soil organic matter and available water capacity. Journal of Soil and Water Conservation. 49: 189-194. Huggins, D.R., Buyanovsky, G.A., Wagner, G.H., Brown, J.R., Darmody, R.G., Peck, T.R., Lesoing, G.W., Vanotti, M.B., and Bundy, L.G. 1998. Soil organic C in the tallgrass prairie-derived region of the corn belt: effects of long term crop management. Soil and Tillage Research 47: 219-234. IPCC. 2007. Climate Change 2007: Synthesis Report. Fourth assessment report of the Intergovernmental Panel on Climate Change. IPCC. 2001. Climate Change 2001: The Scientific Basis. Intergovernmental Panel on Climate Change. Jackson, W. 2002. Natural systems agriculture: a truly radical alternative. Agriculture Ecosystems, and Environment 88: 111-117. Jastrow, J.D. 1987. Changes in soil aggregation associated with tallgrass prairie restoration. American Journal of Botany 74: 1656-1664. Jastrow, J.D. 1996. Soil aggregate formation and the accrual of particulate and mineral- associated organic matter. Soil Biology and Biochemistry 28: 665-676. Jenkinson, D.S., Potts, J.M., Perry, J.N., Barnett, V., Coleman, K., and Johnston, A.E. 161 1995. Trends in herbage yields over the last century on the Rothamsted long-term continuous hay experiment. Journal of Agricultural Science 122: 365-374. Jenkinson, D.S., Poulton, P.R., Johnston, A.E., and Powlson, D.S. 2004. Turnover of nitrogen-15 labeled fertilizer in old grassland. Soil Science Society of America Journal 68: 865-875. JMP, Version 7, SAS Institute Inc., Cary, NC, 1989-2007. Karlen, D.L., Wohllenhaupt, N.C., Erbach, D.C., Berry,, E.C., Swan, J.B., Eash, N.S. and Jordahl, J.L. 1994. Crop residue effects on soil quality following 10-years of no- till corn. Soil and Tillage Research 32: 313-327. Kemper, W.D., Rosenau, R.C., 1986. Aggregate stability and size distribution. In: Klute, A. (Ed.), Methods of Soil Analysis Part 1- Physical and Mineralogical Methods. Soil Science Society of America, Inc., Madison, WI, USA, pp. 425- 442. Klute, A., 1986. Water retention: laboratory methods. In: Klute, A. (Ed.), Methods of Soil Analysis Part 1- Physical and Mineralogical Methods. Soil Science Society of America, Inc., Madison, WI, USA, pp. 635-662. Knapp, A.K. and T.R. Seastedt. 1998. Introduction: Grasslands, Konza Prairie and Long Term Ecological Research. In: Knapp, A.K., Briggs, J.M., Hartnett, D.C., and Collins, S.L. (Eds.). Grassland Dynamics. Oxford University Press, New York, NY, pp. 3-18. Kucharik, C.J. Fayram, N.J. and Cahill, K.N. 2006. A paired study of prairie carbon stocks, fluxes, and phenology: Comparing the worlds oldest prairie restoration with an adjacent remnant. Global Change Biology. 12: 122-139. Kucera, C.L. and Dahlman, R.C. 1968. Root-rhizome relationships in fire-treated stand of big bluestem, Andropogon gerardii. American Midland Naturalist 80: 268- 271. Lal, R. 1994. Methods and guidelines for assessing sustainable use of soil and water resources in the tropics. Soil Management and Support Service, United Stated Department of Agricultlure. SMSS Technical Monograph no. 21. Lal, R. 1997. Degradation and resilience of soils. Philosophical Transactions of the Royal Society 352: 997-1010. Lal, R. 2004. Soil carbon sequestration impacts on global climate change and food security. Science 304: 1623-1627. Lal, R. 2007a. Soils and sustainable agriculture: A review. Agronomy for Sustainable Development 27: 1-8. Lal, R. 2007b. Soil Science and the Carbon Civilization. Soil Science Society of America Journal 71: 1425-1437. Lal, R. 2008. Carbon Sequestration. Philosphical Transactions of the Royal Society, Biological Sciences, 363: 815-830. Lal, R., Kimble, J.M., Follett, R.F., and Cole, C.V. 1999. The Potential of U.S. Cropland to Sequester Carbon and Mitigate the Greenhouse Effect. Boca Raton, FL. CRC Press. Lal, R., Kimble, J.M., Ilvari, T., Sobecki, T.M. 2003. Soil Degradation in the United States. CRC Press, Boca Raton, FL. Lal, R., and Shukla, M. K. 2004. Principles of Soil Physics. Marcel Dekker, Inc., New 162 York, N.Y. Laupwayi, N.Z., Harker, K.N., Clayton, G.W., O’Donovan, J.T., and Blackshaw, R.E. 2009. Soil microbial response to herbicide applied to glyphosate-resistant canola. Agriculture, Ecosystems, and Environment 129: 171-176. Lewandowski, A., Zumwinkle, M., and Fish, A. 1999. Assessing the soil system: A review of soil quality literature. Minnesota Department of Agriculture, Energy and Sustainable Agriculture Program, St. Paul, MN. Lutzow, M. V., Kogel-Knabner, I., Ekschmitt, K., Flessa, H., Guggenberger, G., Matzner, E., and Marschner. B. 2007. SOM fractionation methods: Relevance to functional pools and to stabilization mechanisms. Soil Biology and Biochemistry. 39:2183-2207. Marriott, E.E. and Wander, M. M. 2006b. Total and Labile Soil Organic Matter in Organic and Conventional Farming Systems. Soil Science Society of America Journal 70:950-959. Maly, S., Kralovec, J., and Hampel, D. 2009. Effects of long-term mineral fertilization on microbial biomass, microbial activity and presence of r- and K-strategists. Biology and Fertility of Soils 45: 753-760. Matamala, R., Jastrow, J.D., Miller, R.M., and Garten, C.T. 2008. Temporal changes in C and N stocks and Restored Prairie: Implications for C sequestration strategies. Ecological Applications 18: 1470-1488. McLauchlan. K. 2006. The nature and longevity of agricultural impacts on soil carbon and nutrients: A review. Ecosystems 9: 1364-1382. McLauchlan, K.K., Hobbie, S.E., and Post, W.M. 2006. Conversion from agriculture to grasslands builds soil organic matter on decadal timescales. Ecological Applications 16: 143-153. McQuaid, B.F., and Olson, G.L. 1998. Impact of carbon sequestration on functional I indicators of soil quality. In Lal, R., Kimble, J.M., Follett, R.F., and Stewart, B.A. (Eds) Soil Processes and the Carbon Cycle. CRC Press, Boca Raton, FL. Miller, R. M. and Jastrow, J. D. 2000. Mycorrhizal fungi influence soil structure. In: Kapulnik, Y. and Douds, D. D. (Eds.). Arbuscular Mycorrhizae: Physiology and Function. Kluwer Academic Publishers, the Netherlands, pp. 3-18. Montgomery, D.R. 2007. Dirt: The erosion of civilizations. University of California Press, Berkeley, CA. Moore, J.M., Klose, S., and Tabatabai, M.A. 2000. Soil microbial biomass carbon and nitrogen as affected by cropping systems. Biology and Fertility of Soils 31: 200- 210. Nelson, D.W., Sommers, L.E., 1996. Total carbon, organic carbon, and organic matter. In Sparks, D.L. et al. (Ed.), Methods of Soil Analysis, Part 3, ASA and SSSA, Madison, WI, pp.417-435. Omay, A.B., Rice, C.W., Maddux, L.D., and Gordon, W.B. 1997. Changes in soil microbial and chemical properties under long-term crop rotation and fertiliztion. Soil Science Society of America Journal 61: 1672-1678. Parton, W.J., Brookes, P.C., Coleman, K., and Jenkinson, D.S. 1987. Dynamics of C, N, S, and P in grassland soils: a model. Biogeochemistry 5: 109-131. Paustian, K., Andren, O., Janzen, H. H., Lal, R., Smith, P., Tian, G., Tiessen, H., Van 163 Noordwijk, M., and Woomer, P.L. 1997. Agricultural soils as a sink to mitigate

CO2 emiisions. Soil Use and Management 13: 230-244. Post, W.M. and Kwon, K.C. 2000. Soil carbon sequestration and land-use change: process and potential. Pineiro, G., Jobbagy, E.G., Baker, J., Murray, B.C., and Jackson, R.B. 2009. Set-asides can be a better climate investment than corn ethanol. Ecological Applications 19: 277-282. Potthoff, M., Steenwerth, K.L, Jackson, L.E., Drenovsky, R.E., Scow, K.M., and Joergensen, R.G. 2006. Soil microbial community composition as affected by restoration practices in California grassland. Soil Biology and Biochemistry 38: 1851-1860. Powlson, D.S., Brookes, P.C., and Christensen, B.T. 1987. Measurements of soil microbial biomass provides an early indication of changes in total soil organic matter due to straw incorporation. Soil Biology and Biochemistry 19: 159-164. Reganold, J.P., Elliott, L.F., and Unger, Y.L. 1987. Long-term effects of organic and conventional farming on soil erosion. Nature 330: 370-72. Reganold, J.P., Palmer, A.S., Lockhart, J.C., and Macgregor, A.N. 1993. Soil quality and financial performance of biodynamic and conventional farms in New Zealand. Science 260: 344-349. Reicosky, D.C., Dugas, W.A., and Torbert, H.A., 1997. Tillage-induced soil carbon dioxide loss from different cropping systems. Soil and Tillage Research 41: 105- 118. Rice, C.W., Todd, T.C., Blair, J.M., Seastedt, T.R., Ramundo, R.A. and Wilson, G.W.T. 1998. Belowground Biology and Processes. In: Knapp, A.K., Briggs, J.M., Hartnett, D.C., and Collins, S.L. (Eds.). Grassland Dynamics. Oxford University Press, New York, NY, pp. 244-264. Robertson, G.P., et al. 2008. Sustainable Biofuels Redux. Science 322: 49-50. Rutledge, E.M., Wilding, L.P., and Elfield, M. 1967. Automated particle size separation by sedimentation. Soil Science Society of America Proceedings 31: 287-288. Sa, J.C. de M., Cerri, C. Dick, W.A., Lal, R. Filho, S.P.V. , Piccolo, M.C. and Feigl. B.E. 2001. Organic Matter Dynamics and Carbon Sequestration Rates for a Tillage Chronosequence in a Brazilian Oxisol. Soil Science Society of America Journal 65:1486-1499. SAS Institute, 2002. The SAS System for Microsoft Windows Release 8.2 SAS Institute, Cary, NC. Schimel, D.S. 1995. Terrestrial ecosystems and the carbon cycle. Global Change Biology 1: 77-91. Schmer, M.R., Vogel, K.P., Mitchell, R.B., and Perrin, R.K., 2008. Net energy of cellulosic ethanol from switchgrass. Proceedings of the National Academy of Sciences U.S.A. 105: 464-469. Seastedt, T.R., and Knapp, A.K. 1993. Consequences of non-equilibrium resource availability across multiple time scales: the transient maxima hypothesis. American Naturalist 141: 621-633. Six, J., Conant, R.T., Paul, E.A., and Paustian, K. 2002. Stabilization mechanisms of

164 soil organic matter: implications for C-saturation of soils. Plant and Soil 241: 155-176. Smith, J.S., and Paul, E.A., 1990. The significance of soil microbial biomass estimations. In Bollage, J.M., and Stotzky, G. (Eds) Soil Biochemistry. Marcel Dekker, Inc., New York, NY. Sollins, P., Homann, P., and Caldwell, B.A. 1996. Stabilisation and destabilization of soil organic matter: mechanisms and controls. Geoderma 74: 65-105. Tarnocai, C., Canadell, J.G., Schuur, E.A.G., Kuhry, P., Mazhitova, G., and Zimov, S. 2009. Soil organic carbon pools in the northern circumpolar permafrost region. Global Biogeochemical Cycles 23: GB2023. Tiessen, H., and Stewart, J.W.B. 1983. Particle size fractions and their use in studies of soil organic matter: II. Cultivation effects on organic matter composition in size fractions. Soil Science Society of America Journal 47: 509-514. Tiessen, H., Cuevas, E., and Chacon, P. 1994. The role of soil organic matter in sustaining soil fertility. Nature 317: 783-785. Tilman, D., Hill, J. and Lehman, C. 2006. Carbon-negative biofuels from low-input high diversity grassland biomass. Science 314: 1598-1600. Vance, E.D., P.C. Brookes, and Jenkinson, D.S. 1987. An extraction method for measuring soil microbial biomass C. Soil Biology and Biochemistry19:703-707. Voroney, R.P., Winter, J.P., and Beyaert, R.P. 1993. Soil microbial biomass C and N. p. 277-286. In M.R. Carter (ed.) Soil sampling and methods of analysis. Lewis Publishers, Boca Raton, FL. Wander, M.M., and Bollero, G.A. 1999. Soil Quality Assessment of Tillage Impacts in Illinois. Soil Science Society of America Journal 63: 961-971. Wardle, D.A., and D. Parkinson. 1990. Effects of three herbicides on soil microbial biomass and activity. Plant and Soil 122: 21-28. Wardle, D.A., Yeates, G.W., Watson, R.N., and Nicholson, K.S. 1995. The detritus food-web and the diversity of soil fauna as indicators of disturbance regimes in agroecosystems. Plant and Soil 170: 35-43. Warembourg, F.R. and Esterlich, H.D. 2001. Plant phenology and soil fertility effects on below-ground carbon allocation for an annual (Bromus madritensis) and a perennial (Bromus erectus) grass species. Soil Biology and Biochemistry 33: 1291-1303. Weil, R.R., and Magdoff, F. 2004. Significance of soil organic matter to soil quality and health. In Magdoff, F. and Weill, R.R. (Eds) Soil Organic Matter in Sustainable Agriculture. CRC Press, Boca Raton, FL. Wilson, T.C., Paul, E. A. and Harwood, R.R. 2001. Biologically active soil organic matter fractions in sustainable cropping systems. Applied Soil Ecology 16: 63 Yoder, R.E. 1936. A direct method of aggregate analysis of soils and a study of the physical nature of erosion losses. Journal of the American Society of Agronomy 28: 337-351. Zak, D.R., Holmes, W.E., White, D.C., Peacock, A.D., and Tilman, D. 2003. Plant diversity, soil microbial communities, and ecosystem function: are there any links? Ecology 84: 2042-2050.

165 APPENDIX A

Raw Data from Ohio Prairie Restoration Project

166 Treatment Depth % C % N %WSA MWD Bulk Density P951 10 3.71 0.34 95.59 5.07 1.24 P952 10 2.95 0.25 96.52 4.75 1.37 P953 10 2.66 0.23 96.71 5.19 1.29 P954 10 2.38 0.21 96.76 4.97 1.38 P951 20 2.75 0.23 95.14 4.55 1.43 P952 20 1.98 0.17 95.63 4.46 1.54 P953 20 1.60 0.16 93.62 4.18 1.54 P954 20 1.89 0.17 96.53 4.63 1.49 P951 30 2.72 0.30 95.25 4.48 1.40 P952 30 2.18 0.20 94.51 3.76 1.37 P953 30 1.30 0.13 93.37 3.61 1.54 P954 30 1.78 0.16 95.1 4.03 1.50 P951 40 2.54 0.22 94.82 4.17 1.30 P952 40 1.81 0.17 90.39 5.26 1.56 P953 40 0.93 0.11 87.85 2.12 1.63 P954 40 1.39 0.14 88.5 2.52 1.53 P001 10 3.08 0.28 95.48 1.16 1.21 P002 10 3.05 0.24 95.15 4.66 1.34 P003 10 2.67 0.21 95.26 4.91 1.36 P004 10 2.64 0.21 96.03 4.77 1.25 P001 20 2.35 0.22 96.84 4.86 1.49 P002 20 2.41 0.21 95.74 4.10 1.45 P003 20 2.36 0.19 95.46 5.05 1.53 P004 20 2.36 0.19 94.09 3.89 1.48 P001 30 2.33 0.20 68.3 2.59 1.41 P002 30 2.40 0.20 96.71 4.90 1.41 P003 30 2.17 0.18 93.44 4.41 1.57 P001 40 2.01 0.23 93.91 3.30 1.39

167 P002 40 1.02 0.09 92.66 3.88 1.63 P003 40 2.40 0.19 93.07 3.84 1.42 P771 10 3.68 0.26 97.25 5.12 1.19 P772 10 4.14 0.27 96.73 5.11 1.03 P773 10 2.81 0.21 97.31 5.19 1.15 P774 10 3.15 0.23 96.69 5.12 1.08 P771 20 2.32 0.18 96.76 4.90 1.43 P772 20 1.89 0.16 96.73 4.92 1.52 P773 20 1.67 0.14 97.02 5.00 1.42 P774 20 1.55 0.13 96.89 4.93 1.52 P771 30 1.90 0.15 93.82 4.64 1.45 P772 30 1.90 0.15 97.13 5.00 1.52 P773 30 1.65 0.14 95.61 4.67 1.44 P774 30 1.46 0.13 95.23 4.74 1.52 P771 40 2.35 0.19 95.72 4.40 1.36 P772 40 2.10 0.16 96.03 4.62 1.47 P773 40 1.19 0.10 96.24 4.69 1.56 P774 40 1.53 0.14 95.14 4.52 1.59 AG1 10 1.75 0.17 68.92 0.89 1.45 AG2 10 2.26 0.22 83.53 2.70 1.52 AG3 10 1.64 0.17 45.51 1.00 1.47 AG4 10 2.93 0.27 86.47 2.54 1.41 AG1 20 1.97 0.17 93.49 3.70 1.62 AG2 20 2.07 0.20 87.26 2.15 1.62 AG3 20 1.30 0.14 56.81 0.42 1.65 AG1 30 1.80 0.17 77.46 2.17 1.59 AG2 30 2.18 0.21 86.27 1.65 1.59 AG3 30 1.26 0.14 72.06 0.90 1.52 AG1 40 0.70 0.10 58.42 1.64 1.56

168 AG2 40 1.28 0.15 90.54 2.91 1.58 AG3 40 1.11 0.13 87.58 1.62 1.48 AG4 40 1.64 0.18 91.81 3.19 1.55 LA1 10 1.82 0.17 81.74 2.60 1.40 LA2 10 1.87 0.18 88.77 3.77 1.30 LA3 10 1.67 0.17 87.27 3.93 1.23 LA4 10 1.83 0.18 87.23 3.85 1.51 LA1 20 1.66 0.22 74.62 1.33 1.56 LA2 20 1.58 0.15 70.56 1.87 1.56 LA3 20 1.34 0.15 78.18 2.88 1.58 LA4 20 1.59 0.17 82.97 2.97 1.59 LA1 30 1.46 0.14 80.16 1.09 1.50 LA2 30 1.59 0.17 76.31 2.09 1.56 LA3 30 1.20 0.14 73.97 2.25 1.55 LA4 30 1.57 0.17 82.36 2.68 1.59 LA1 40 1.39 0.14 82.90 0.97 1.52 LA2 40 1.36 0.15 85.69 1.41 1.51 LA3 40 0.82 0.11 82.82 1.47 1.52 LA4 40 1.13 0.14 86.10 1.22 1.53

169 Total POM-C Treatment Depth AWC Porosity POM-N P951 10 0.10 0.47 7.10 0.50 P952 10 0.11 0.45 5.79 0.36 P953 10 0.19 0.48 6.43 0.42 P954 10 0.15 0.44 4.52 0.32 P951 20 0.11 0.46 2.16 0.17 P952 20 0.09 0.38 2.33 0.18 P953 20 0.10 0.36 2.34 0.17 P954 20 0.10 0.38 2.25 0.16 P951 30 0.06 0.43 2.55 0.19 P952 30 0.10 0.43 2.52 0.19 P953 30 0.12 0.38 2.88 0.20 P954 30 0.09 0.38 1.83 0.15 P951 40 0.10 0.47 2.39 0.18 P952 40 0.08 0.41 1.38 0.10 P953 40 0.06 0.34 1.24 0.10 P954 40 0.06 0.36 1.49 0.11 P001 10 0.20 0.50 5.66 0.41 P002 10 0.15 0.45 6.60 0.47 P003 10 0.13 0.45 6.72 0.45 P004 10 0.18 0.49 5.88 0.39 P001 20 0.08 0.39 3.45 0.24 P002 20 0.12 0.43 2.69 0.20 P003 20 0.10 0.41 3.63 0.27 P004 20 0.11 0.40 3.23 0.21 P001 30 0.10 0.44 2.59 0.19 P002 30 0.15 0.46 2.76 0.21 P003 30 0.07 0.39 2.65 0.19

170 P004 30 0.10 0.39 2.64 0.20 P001 40 0.13 0.44 2.83 0.20 P002 40 0.06 0.38 1.78 0.14 P003 40 0.11 0.45 2.94 0.18 P771 10 0.21 0.54 11.40 0.53 P772 10 0.19 0.58 16.45 0.78 P773 10 0.28 0.57 8.66 0.47 P774 10 0.12 0.54 19.85 1.05 P771 20 0.10 0.42 4.99 0.26 P772 20 0.12 0.38 3.35 0.16 P773 20 0.13 0.41 3.96 0.20 P774 20 0.13 0.38 3.72 0.22 P771 30 0.10 0.42 2.31 0.14 P772 30 0.08 0.36 3.31 0.17 P773 30 0.12 0.40 2.29 0.13 P774 30 0.36 2.96 0.16 P771 40 0.05 0.41 2.32 0.15 P772 40 0.15 0.41 1.98 0.14 P773 40 0.08 0.35 1.57 0.10 P774 40 0.04 0.32 1.74 0.13 AG1 10 0.12 0.38 2.33 0.19 AG2 10 0.09 0.29 2.72 0.21 AG3 10 0.14 0.41 2.88 0.23 AG4 10 0.10 0.42 5.25 0.38 AG1 20 0.10 0.34 2.03 0.15 AG2 20 0.06 0.35 1.90 0.15 AG3 20 0.11 0.35 1.99 0.18 AG4 20 0.10 0.44 2.38 0.20 AG1 30 0.13 0.38 1.47 0.12

171 AG2 30 0.09 0.39 1.43 0.11 AG3 30 0.16 0.42 1.64 0.13 AG4 30 0.09 0.43 1.42 0.11 AG1 40 0.10 0.37 0.86 0.08 AG2 40 0.07 0.37 0.89 0.09 AG3 40 0.12 0.42 1.12 0.09 AG4 40 0.07 0.42 1.35 0.11 LA1 10 0.15 0.36 3.90 0.23 LA2 10 0.17 0.43 3.86 0.24 LA3 10 0.14 0.40 4.17 0.27 LA4 10 0.12 0.36 4.52 0.31 LA1 20 0.13 0.33 1.98 0.13 LA2 20 0.15 0.36 2.06 0.15 LA3 20 0.10 0.33 2.50 0.17 LA4 20 0.10 0.32 2.47 0.17 LA1 30 0.13 0.34 1.13 0.08 LA2 30 0.13 0.33 1.81 0.13 LA3 30 0.12 0.32 1.67 0.13 LA4 30 0.11 0.34 1.75 0.13 LA1 40 0.12 0.35 1.29 0.10 LA2 40 0.11 0.35 0.93 0.08 LA3 40 0.10 0.33 1.15 0.08 LA4 40 0.09 0.34 1.20 0.10

172