Using Water Quality Trading to Promote Conservation Measure

USING WATER QUALITY TRADING TO PROMOTE CONSERVATION MEASURE

ADOPTION IN THE BLANCHARD RIVER WATERSHED, OHIO, IN THE CONTEXT OF

CLIMATE CHANGE

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Doctor of Philosophy in the Graduate School of The Ohio State University

Yanting Guo, M.Sc.

Environmental Science Graduate Program

The Ohio State University

2018

Dissertation Committee:

Dr. Richard Moore

Dr. Karen Mancl

Dr. Charles Goebel

Dr. Eric Toman

Copyrighted by

Yanting Guo

2018

ABSTRACT

Agricultural phosphorus loading has been identified as the main cause of Lake Erie

eutrophication and Harmful Algae Blooms since the mid-1990s. Efforts for alleviating the problem have been focused on promoting the adoption of conservation measures, such as the

best management practices (BMPs). Water quality trading (WQT), a market-based mechanism that allows one pollution source (e.g. a factory) to meet their regulatory obligations by using pollutant reductions created by another source, such as agriculture, is a method to promote conservation through the lower pollution remediation costs of that source. Agriculture has a vast potential for supplying low cost conservation, however, limited studies have been done on farmers’ opinions about WQT or their selection and adoption of conservation measures in WQT programs. This study aims to fill this knowledge gap. The general objective of this study is to evaluate whether WQT has the potential to serve as an incentive to promote water quality conservation measures in the Western Lake Erie Basin of Ohio in the USA. Based on the case study of the Blanchard River watershed which is a subwatershed of the Maumee River, the largest tributary flowing into the Western Lake Erie, this study investigated 1) how WQT models affected BMP selection and adoption, as well as farmers’ willingness to participate in WQT, in order to inform the current debate over credit “stacking” (trading more than one type of pollutant credits in one trading project) with insights from the potential credit suppliers; 2) the possibility of expanding the use of WQT from agriculture to another often-neglected nonpoint source, rural septic systems, encouraging rural households to improve nutrient removal efficiency of their systems; and 3) how farmers select and adopt BMPs based on their observation of climate change and prediction of its impact on local water quality.

With quantitative and qualitative data collected from an in-person questionnaire survey

completed by 96 farmers and 88 nonfarmers in the Blanchard River watershed, I employed binary logistic regression, special analysis and content analysis to investigate the different aspects of people’s perceptions about WQT and their adoption of conservation measures in

WQT and climate change scenarios. The major findings are as follows.

First, the farmers showed a clear preference for the credit “stacking” trading model

(termed “All-in-One” WQT model) over the conventional single credit trading model. This preference led to enhanced interest in participating in WQT. Farmers valued both the economic and ecological benefits represented by the “All-in-One” model. Although planting cover crops was the most popular BMPs for farmers who intended to adopt in a WQT scenario, farmers’ adoption of cover crops or other BMPs was not significantly associated with their preference for trading model. Factors associated with farmers’ BMP adoption were much more complicated and were inconsistent among locations, types of BMPs and adoption stages. For farmers in the Blanchard River watershed, previous experience is critical to cover crop adoption while income and land tenure is important to no-till plowing adoption. In general, when considering BMP adoption, farmers were most concerned about yield loss and improvement in water quality. Given the complexity and heterogeneity inherent to BMP adoption, as well as the stewardship valued by farmers in the Blanchard River watershed, a small-scale, community- based WQT project might be more likely to succeed. Second, 58% of households that used septic systems were found to be interested in participating in WQT. WQT as an incentive for septic system upgrades had higher levels of acceptance in certain locations, namely upstream households of Blanchard River and Lye Creek. These households were more concerned about

the environment, perceived the local water quality to be degraded and were aware of the

limitation of their septic systems. Pilot WQT projects should be focused on approaching these

households. Third, the study found that farmers’ observation of changes in climate was the

main driver of their action. Farmers who had already taken action were those who observed

climate trends more accurately. However, the observation did not significantly affect farmers’ intentions to adopt additional conservation measures in the future, even for those who had already taken action. Consistent with the farmers’ willingness to participate in WQT, environmental concerns and income were the most important factors in BMP adoption in the climate change scenario.

In sum, while many farmers and septic system users in the Blanchard River watershed were open to the idea of WQT, some groups were more interested than others. Environmental awareness, income level, stewardship values, and concern about local water quality, were the indicators that could help WQT project designers identify potential participants.

iii

Dedicated to my husband Li and children Charlotte and Andrew, who supported and

accompanied me during my Ph.D. journey

ACKNOWLEDGEMENT

This dissertation would not have been possible without the guidance, encouragement,

and unreserved support from my advisor, Dr. Richard Moore. In the past 4 years, I am indebted

to him for his teaching, insights, and ceaseless curiosity. I truly appreciate the support I received

from my committee. I would like to thank Dr. Karen Mancl, who set an example, in academia

and beyond, for a female Ph.D. student like me. I also appreciate her willingness to step in as

committee chair. I feel grateful to Dr. Charles Goebel for his generous support, especially in the

last 2 years. I also thank Dr. Eric Toman for his inspiration in my study of the social dimensions of climate change issues.

Special thanks go to the famers and residents who helped me complete the survey in

Hancock County, I am very grateful to the valuable opinions and the kindness they shared with me and my family. I am also thankful for Phil Martins of the Blanchard River Watershed

Partnership, Dr. Ed Lentz at OSU Extension of Hancock County, Drew Humphreys at Hancock

County Natural Resources Conservation Service, Dr. Gary Wilson at Hancock County Farm

Bureau and Dr. Tim Murphy at University of Findlay, who had helped me from design to delivery of the survey. I would like to thank the three survey assistants from University of

Findlay, Mishael Theis, Drew Garverick and Eli deNijs. In addition, I am thankful for Josh Sadvari at the OSU Research Common, who provided technical support in GIS. Last, I want to thank the strangers who pulled our car out from a ditch on a snowy evening during survey field work.

I am very appreciative to the various funding sources provided for my education. First,

the Fay Fellowship supported my first year of study. Then I was funded by the USDA-

Agriculture and Food Initiative: Grant No. 416-40-63C followed by funding provided by the OSU

Targeted Investment in Excellence (TIE) initiative.

I also appreciate my friend Zhe Zhang, who always lent me a hand when I needed it;

Yina Xie for sharing her experience and friendship with me; Long Lin for her warm help and encouragement.

Finally, I am most grateful to my husband Li Zhang, who has been my best friend and mentor for 17 years. Without his love, support and sacrifice, I would not have been able to pursue and achieve my goals. I am also thankful to my daughter Charlotte for her company during the survey, her understanding when I needed to study, and for all the joy she brings to the family. I am grateful to my son Andrew, who has been a patient baby and a wonderful sleeper, allowing me to finalize the dissertation soon after he was born. Also thanks go to my parents who were so supporting during my education. Lastly, I want to thank my parents-in-law who came all the way from China to support our family when I was finishing this dissertation.

VITA

2002-2006…………………………… B.S. Biological Sciences, Zhejiang University

2006-2007…………………………… M.Sc. Biodiversity, Conservation and Management, University of

Oxford

2013-2014…………………………… ESGP Fay Fellow, The Ohio State University

2014-2015…………………………… ESGP Graduate Administrative Associate, The Ohio State

University

2015-2017…...……………………… Graduate Research Associate, School of Environment and Natural

Resources, The Ohio State University

PUBLICATIONS

Guo, Yanting. 2017. Water Quality Trading for On-site Septic System Nutrient Management in Lake Erie Watershed under the Changing Climate Conditions. Poster presented at the 2017 OARDC Annual Conference. Columbus, OH.

Guo, Yanting. 2015. A Feasibility Study of Water Quality Trading in the Former Great Black Swamp of the Lake Erie Watershed under Changing Climate Conditions in the Corn Belt. Poster presented at the National Worship on Water Quality Markets. Lincoln, NE.

Guo, Yanting. 2015. Application of Water Quality Trading in the former Great Black Swamp under Changing Climate Conditions of Lake Erie. Poster presented at the 1st ESGP Poster Symposium, Columbus, OH.

Xu, J., B. Gu, Y. Guo, J. Chang, Y. Ge, Y. Min, X. Jin. 2010. "A Cellular Automata Model for Population Dynamics Simulation of Two Plant Species with Different Life Strategies." Proceedings of 2010 IEEE International Conference of Intelligent Systems and Knowledge Engineering, Nov. 15-16, 2010, Hangzhou, China: 517-523.

vii

Chang, J., J. Xu, Y. Ge, B. Gu, X. Jin, Y. Guo, Y. Min. 2007. "Studies of Population Quantity Dynamics of an Endangered Plant Based on Cellular Automata." Mind and Computation 1(4): 465-474.

Guo, Yanting. and Gabriel H. Calvi. 2001. "Investigation and Expectation of Aquatic Breeding in Southern China Coast ---- From the Studies of Marine Ecology in the ‘Marine Fishing Village’ at Shen'ao Bay, Nan'ao Island and Zhelin Bay at Raoping, Guangdong, China." Ecological Science (Chinese) 20(4): 94-98.

Field of Study

Major Field: Environmental Science

viii

TABLE OF CONTENTS

Abstract …………………………………………………………………………………………… i

Acknowledgments …………………………………………………………………………………………… v

Vita …………………………………………………………………………………………… vii

Table of Contents …………………………………………………………………………………………… ix

List of Tables …………………………………………………………………………………………… xii

List of Figures …………………………………………………………………………………………… xiii

Chapter 1. Introduction …………………………………………………………………………………………… 1

1.1 Background of Study …………………………………………………………………………………………. 1

1.2 What is Water Quality Trading .…………………………………………………………………………… 3

1.3 Debates over Water Quality Trading …………………………………………………………………… 5

1.4 Recent Advances of Water Quality Trading ………………………………………………………… 12

1.5 Motivation of Study ………………………………………………………………………………………….. 14

1.6 Objectives of Study ………………………………………………………………………………………….. 15

Chapter 2. “All-in-One” Credit Stacking Water Quality Trading Model and BMP Adoption in the Blanchard River Watershed of Ohio……………………………………………………. 26

2.1 Introduction …………………………………………………………………………………………… 26

2.2 Methodology …………………………………………………………………………………………… 34

2.2.1 Study Area …………………………………………………………………………………………… 34

2.2.2 Survey …………………………………………………………………………………………… 35

2.2.3 Analytical Approach ….…………………………………………………………………………………… 37

2.2.4 Limitations of Methodology ..………………………………………………………………………… 39

2.3 Results …………………………………………………………………………………………… 41

2.3.1 Characteristics of Farmers and Farms …………………………………………………………… 41

2.3.2 Current Adoption of BMPs …………………………………………….……………………………… 43

2.3.3 Farmers’ Preference for WQT Models: “All-in-One” Credit Stacking Model vs. Single Credit Model…………………………………………….……………………………………………………….. 46

2.3.4. Would All-in-One Trading Model Encourage Cover Crops and No-Till Adoption? ………………………………….……………………………………………………………………….……….. 47

2.3.5 Farmers’ Interest in WQT ……….…………………………….…………………………….…………. 53

2.4 Discussion …………………………………………………………………………………………… 58

2.5 Conclusion …………………………………………………………………………………………… 61

Chapter 3. Water Quality Trading for On-site Septic System Nutrient Management under the Changing Climate Conditions…………………………………………….…………………………. 71

3.1 Introduction …………………………………………………………………………………………… 71

3.2 Methodology …………………………………………………………………………………………… 78

3.2.1 Study Area …………………………………………………………………………………………… 78

3.2.2 Data Collection …………………………………………………………………………………………… 79

3.2.3 Analytical Approach…………………………………………….…………………………………………. 81

3.3 Results and Discussion…………………………………………….…………………………………………… 83

3.3.1 Current Status of Septic Systems in the Blanchard River Watershed…………………………………………….………………………………………………………………… 83

3.3.2 Perceived Environmental Quality and Risk by Households in the Blanchard River Watershed…………………………………………….……………………………………………………………. 87

3.3.3 Households’ Willingness to Upgrade Septic Systems ……………………………………… 90

3.3.4 Factors Associated with Households’ Willingness to Upgrade Septic Systems… 98

3.4 Conclusion …………………………………………………………………………………………… 102

Chapter 4. Climate Change and BMP Adoption in the Blanchard River Watershed, Ohio …………………………………………….……………………………………………………………………………… 113

4.1 Introduction …………………………………………………………………………………………… 113

4.2 Methodology …………………………………………………………………………………………… 117

4.2.1 Data Collection …………………………………………………………………………………………… 117

4.2.2 Data Analysis …………………………………………………………………………………………… 118

4.3 Results …………………………………………………………………………………………… 121

4.3.1 Farmers’ Observation of Climate Change………………………………………………………. 121

4.3.2 Farmers’ Action to Address Climate Change…………………………………………………… 127

4.3.3 Farmer’s Intention to Adopt Additional BMPs under Future Climate Conditions……………………………………………….…………………………………………………. 130

4.4 Discussion …………………………………………………………………………………………… 134

4.5 Conclusions …………………………………………………………………………………………… 138

Chapter 5. Conclusion …………………………………………………………………………………………… 144

Bibliography …………………………………………………………………………………………… 154

Appendix A: Survey Questionnaire……………………………………………………………………………….. 182

LISTS OF TABLES

Table 2.1 Responses of farmers’ survey ……………………………………………………………………….. 36

Table 2.2 Variables associated with farmers’ interested in WQT..………………………………… 38

Table 2.3 Current adoption rate of BMPs ……………………………………………………………………. 45

Table 2.4 Results of binary logistic regression (presented in odds ratio) ……………………… 52

Table 3.1 Principal reasons for septic system failure (adopted from ODH & Ohio EPA, 2013) ………………………………………………………………………………………………………………………….. 74

Table 3.2 Responses of septic system survey, from both farming and non-farming households…………………………………………………………………………………………………………………… 81

Table 3.3 Variables associate with households' interest in septic system upgrades ……… 82

Table 3.4 Soil types and natural drainage class at the locations of the responded households………………………………………………………………………………………………………………….. 86

Table 3.5 Households’ interests in septic system upgrades under the scenarios of three WQT models………………………………………………………………………………………………………………… 94

Table 3.6 Results of binary logistic regression for households’ willingness to upgrade septic systems in three scenarios (presented in odds ratio) ………………………………………… 99

Table 4.1 Variables potentially associated with farmers’ action and intention to adopt additional BMPs………………………………………………………………………………………………………….. 119

Table 4.2 Binary logistic regression results for farmers’ action and associated variables (presented in odds ratio)…………………………………………………………………………………………….. 129 Table 4.3 Binary logistic regression results for farmers’ action and observation of rain, flood and temperature (presented in odds ratio)…………………………………………………………. 130

Table 4.4 Binary logistic regression results for farmers’ intention to adopt additional BMPs and associated variables (presented in odds ratio)…………………………………………….. 132

xii

LISTS OF FIGURES

Figure 2.1 Distribution of Farmers' Age (in years)………………………………………………………… 41

Figure 2.2 Distribution of household farming income (in US dollar)..…………………………… 42

Figure 2.3 The role of farming in total household income …………………………………………… 43

Figure 2.4 Type of tillage in-use…………………………………………………………………………………… 44

Figure 2.5 Farmers’ use of tillage types ……………………………………………………………………… 44

Figure 2.6 Farmers’ preference for trading models……………………………………………………… 47

Figure 2.7 Selection of BMPs for farmers who preferred the “All-in-One” Model………… 51

Figure 2.8 Comparison of farmers with different trading model preferences regarding issues about BMP adoption………………………………………………………………………………………… 52

Figure 2.9 Comparison of farmers with different trading model preferences regarding issues about WQT ……………………………………………………………………………………………………… 57

Figure 2.10 Farmers’ trust in agencies for WQT …………………………………………….…………… 58

Figure 3.1 Pollutant removal soil depths for wastewater infiltrating unsaturated soil (adopted from Mancl and Slater, 2013) ……………….…………………………….………………………… 74

Figure 3.2 PmA (Pewamo silty clay loam) is the most common soil type in Hancock County, accounting for 24.4% of the total area of Hancock County………………….…………… 76

Figure 3.3 (a) Blanchard River Watershed (adopted from USDA, 2011); (b) Study area… 79

Figure 3.4 Septic systems installation year……………….…………………………….……………………. 84

Figure 3.5 Distribution of households using septic systems (the darker color shows older systems) ……………….…………………………….…………………………….…………………………….… 85

Figure 3.6 Maintenance of septic systems in the past 5 years……………….……………………… 87

Figure 3.7 Perceived sewage removal effectiveness……………….…………………………….……… 88

Figure 3.8 Perceived bacteria/virus/pathogen removal effectiveness……………….…………. 89

Figure 3.9 Perceived nutrients removal effectiveness……………….…………………………….…… 89

xiii

Figure 3.10 Households’ willingness to upgrade septic systems in (a) the scenario of intensified rainfall and (b) the scenario of 2015 regulation.……………….………………………… 91

Figure 3.11 Households’ concern about regulation regarding septic systems………………. 92

Figure 3.12 Households’ willingness to upgrade septic systems under three scenarios… 93

Figure 3.13 Geographical distribution of households’ willingness to upgrade septic systems in (a) the intensified rainfall scenario, (b) the 2015 regulation scenario and (c) the WQT scenario ……………….…………………………….…………………………….………………………….. 95

Figure 3.14 Geographical distributions of households’ general interest in WQT (a), WQT with an annual payment (b) and WQT with professional management plan (c)…… 97

Figure 4.1 Farmers’ observation of rain intensity change since the year 2000 …………….. 122

Figure 4.2 Farmers’ observation of rain frequency change since the year 2000……………. 122

Figure 4.3 Elevation of the Blanchard River Watershed (Adopted from NRCS Rapid Watershed Assessment - Data Profile Blanchard River Watershed, 2008) ……………..…… 123

Figure 4.4 Farmers’ observation of flood intensity change since the year 2000 …………… 124

Figure 4.5 Farmers’ observation of flood frequency change since the year 2000 ………… 124

Figure 4.6 Crest Height of the Blanchard River at the City of Findlay 1910-2017 (Data from NOAA National Weather Service) …………………………………………….………………………… 125

Figure 4.7 Farmers’ observation of summer temperature change since the year 2000… 126

Figure 4.8 Farmers’ observation of summer precipitation change since the year 2000… 126

Figure 4.9 Farmers’ prediction of Lake Erie water quality change under future climate conditions…………………………………………………………………………………………………………………… 127 Figure 4.10 Farmers’ willingness to adopt additional BMPs under future climate conditions……………………………………………….…………………………………………………………………. 131

xiv

Chapter 1. Introduction

1.1 Background of Study

The western part of Lake Erie has suffered from increasingly severe eutrophication and frequent harmful algae blooms (HAB) since the mid-1990s (Kane et al., 2014). The toxic or potentially toxic cyanobacterial HABs have threatened the aquatic life as well as the health of millions of people who depend on Lake Erie as a drinking water sources (Michalak et al., 2013;

Stumf et al., 2012). The enlarging hypoxia, also an outcome of eutrophication, has harmed the entire ecosystem of Lake Erie (Arend et al., 2011; Scavia et al., 2014). Since the mid-1990s,

HAB’s have been mainly caused by the overloading of phosphorus, especially the increase of dissolved reactive phosphorus (DRP), most of which was from the Maumee River during springtime (Ho et al., 2017). The intensively managed agriculture of the Maumee River watershed has been a major contributor to the elevated phosphorus loading in Lake Erie

(Cousino et al., 2015; Kalcic et al., 2016; Michalak et al., 2013; Stumpf et al., 2012). About 76% of the Maumee watershed is covered by row crop cultivation (Bosch et al., 2014), which consumes large amounts of fertilizer. The fertilizer utilization rate of most crops is about 1/2 to

1/3 and the remainder has the potential to become runoff in heavy rain events (Tilman et al.,

2001), ending up accumulating in the receiving waterbody.

With the precipitation pattern changes due to climate change, phosphorus loading is expected to increase in the western Lake Erie basin. The Third National Climate Assessment reported that, in the northern part of the U.S., the rainfall amount in very heavy rain events has

increased 30-39% from 1958 to 2012 and will continue to increase by up to 5 times (under RCP

8.5 scenario) by 2081 to 2100 (compared with 1981-2000) (Walsh et al., 2014). Bosch et al.

(2014) predicted that the total phosphorus loading in the western Lake Erie Basin will increase 4%

(under the moderate climate scenario) to 6% (under the pronounced climate scenario) if no action is taken to reduce runoff. To alleviate eutrophication under the future climate conditions, vigorous adoption of agricultural conservation measures is required (Kalcic et al., 2016; Ohio

Department of Agriculture et al., 2013; Ohio EPA, 2010). Particularly, to effectively reach the goal of reducing DRP, a phosphorus species considered most relevant to HAB, more ambitious conservation measure adoption schemes are necessary (Bosch et al., 2014; Cousino et al., 2015;

Kalcic et al., 2016).

It is also notable that a large part of the current agricultural lands in the Maumee River watershed was converted from the former Great Black Swamp, one of the largest wetlands in the U.S., over a century ago. Much of the Black Swamp was drained according to the law passed in 1859 by the Ohio General Assembly authorizing county commissioners to construct drainage ditches (Levy, 2017). The removal of the natural nutrient filter as well as the poorly-drained soil

and deep ditches left behind also contribute to HAB’s in Lake Erie. In the Blanchard River

watershed, one of the subwatersheds of the Maumee, the situation is even worse because

frequent flooding there has caused both water quantity and quality problems. In this particular

area, drainage is the top priority for crop growers. Rather than holding the water on site, the

drainage system collects the rain water on the farmland and quickly directs the water to the

drainage ditches and eventually to rivers. As a result, the downstream area floods more quickly

than in the Black Swamp Era and more nutrients are transported downstream.

Therefore, reducing phosphorus runoff from agricultural lands is the key to ameliorating the

HABs in Lake Eire. A series of best management practices (BMPs) were recommended to farmers by the Phosphorus Task Force of Ohio to reduce the loss of phosphorous (Ohio EPA,

2010 and Ohio Department of Agriculture, et al., 2013). However, as the adoption of BMPs remains voluntary in Ohio, continuous efforts to promote BMPs are of critical importance.

Water quality trading, which is an incentive for nonpoint sources to adopt BMPs, has received more and more attention in the Maumee River watershed. Most recently, the Great Lakes

Commission (GLC), a multi-state organization that promotes conservation of Great Lakes resources, received a grant of $400,000 from the U.S. Department of Agriculture (USDA) to launch a three-year program to test water quality trading as a tool for improving water quality in the western Lake Erie Basin. Meanwhile, to improve the quality of flow data (currently continuous flow monitor is lacking) in the Maumee watershed, in June of 2013, the Ohio

Department of Natural Resources (ODNR) signed an agreement with the United States

Geological Survey (USGS) to install seven additional water quality monitoring stations in the

Maumee River basin—mostly near the mouths of the major tributaries to the Maumee River.

These monitoring stations will give better subwatershed data going forward. The monitoring data is available to download at the National Center for Water Quality Research at the

Heidelberg University (https://ncwqr.org/monitoring/data/).

1.2 What is Water Quality Trading?

The idea of water quality trading (WQT) is based on the concept of an ecosystem services market, in which the benefits provided by nature, or more specifically, the functioning

ecosystems, could be measured economically and assigned a monetary value; therefore, these

benefits, or in other words, ecosystem services, could be traded just like other commodities in

markets. One purpose of assigning a value to nature is to inform people, especially decision- makers that the services provided by nature are not free and nature’s services should not be

used for free. However, to actually get people to pay for using nature is another issue. Until

recently taxes and subsidies have been a major form of payment to encourage conservation

behaviors, but in the past few decades, markets have been developed or proposed that allow

“credits” to be generated by ecosystem services for improvements in water quality, carbon

sequestration, wetlands restoration, etc. (Marshall and Weinberg, 2012 ). So far, several types

of ecosystem services markets are present in the U.S.: carbon markets, water markets,

biodiversity markets, and “stacking” or “bundled” markets that combine carbon, water and

biodiversity markets (Carroll & Jenkins, 2013).

Water quality trading is a market-based mechanism that aims to improve water quality in a

way that maximizes economic efficiency and conserves environmental integrity (Selman et al.,

2009). It is used as one of the tools to achieve the goals of Clean Water Act, which aims to

“restore the nation’s waters to levels that would support fishing, swimming, and the other designated uses on which we rely” (USEPA, 2002). In practice, WQT provides point source dischargers, which usually have a Clean Water Act regulatory requirement to meet (e.g.,

National Pollutant Discharge Elimination System (NPDES) Permit), the flexibility on how to meet these requirements. The dischargers may purchase water quality improvements in the form of credits from another party (which could be either point source or nonpoint source) with lower pollution control cost instead of or in addition to installing onsite technology (Willamette

Partnership, World Resources Institute, and the National Network on Water Quality Trading,

2015). WQT can take place between point and nonpoint sources, two point sources, or two

nonpoint sources. The trading between point and nonpoint sources is of special interest

because the nonpoint sources are usually not regulated by current legislation. The trading then

serves as a substitute to a regulatory obligation that motivates pollution reduction actions.

Similar to other ecosystem services trading, WQT projects draw multilateral collaboration of

different stakeholders. The essential stakeholders are the regulator (e.g., the U.S.

Environmental Protection Agency), the practitioner (e.g., NGOs), the credit buyer (e.g., point sources), and the credit supplier (e.g., another point source or non-point source). Sometimes a

third party credit certification organization is also involved (e.g., the Willamette Partnership).

The accommodation of the interests of different stakeholders is usually the key to a successful

WQT trading project.

1.3 Debates over Water Quality Trading

Despite the great potential of WQT discussed above, this approach remains controversial

and is considered risky by many people, especially when nonpoint sources are involved. The

risks are mainly introduced by the inherent uncertainties of WQT. These risks include 1) the

biophysical and scientific uncertainties: how to accurately measure and predict the

performance the conservation measures; 2) extreme event uncertainties: how to control the

variation of conservation measure performance during flood, drought, and other extreme or

stochastic events; 3) behavioral uncertainties: since the nonpoint sources are not usually

regulated by law, and most WQT are multi-year projects, how a WQT project guarantees

sufficient binding force to maintain consistent delivery of credits from its credit suppliers; 4)

regulatory uncertainties: since WQT is a relatively new mechanism, current policies about WQT

could potentially change in the future, and possibly affect credit generation and sales; and 5)

market uncertainties: WQT is a market-based mechanism, but since no mature, universal

market exists for water quality credits, the availability of credit supply and demand, as well as

the credit price are subject to change.

In addition to the five aspects above, there are two other controversial issues related to

WQT. One issue is credit stacking, a design aimed to advance the cost-efficiency of WQT. The

other issue is the selection of a baseline, which is now ambiguously defined and inconsistent

among projects.

Credit stacking has been defined as “selling credits representing two or more spatially

overlapping ecosystem services as separate commodities, each compensating for different permitted impacts” (Robertson et al., 2014). It is an attempt to make WQT even more economically efficient. The rationale for credit stacking is that different ecosystem services can be spatially overlapped, so the credits representing different types of ecosystem services can

also be separately stacked. Different types of credits can be sold separately to different buyers

to compensate for different environmental impacts (Robertson et al., 2014). By selling different types of credits, the cost of WQT can be reduced. Although it is still under debate, many

regulators and practitioners who want to promote ecosystem service markets believe stacking

is a conservation tool with great potential. From an ecological point of view, stacking is a

promising start to value the ecosystem for the multiple services it provides. Stacking allows

credit suppliers to earn extra revenue by selling multiple types of credits; it also lowers the risk

for the situation where a particular conservation measure might fail to produce one type of

credit, as it could be compensated by other types of credits. To take advantage of credit

stacking, suppliers need to install conservation measures that are able to produce multiple

types of credits; in other words, they have to restore a more complex ecosystem that has

multiple functions and provides multiple services. Therefore, compared with the conventional

single credit trading, stacking has the potential to incorporate more ecosystem structures and

functions leading to biocomplexity.

The term “All-in-One” trading model in this dissertation refers to the concept of “stacking”.

Robertson et al. (2014) summarized that this trading model can be applied to the situations 1)

where a single spatial unit as a whole can generate multiple credit types, which can be sold

separately (Willamette Partnership, 2010); 2) within a spatial unit where different credit types

are generated in different subunits that are not spatially overlapping or lineated (NCEEP, 2009);

or 3) where multiple ecosystem services generated within one property are bundled and sold

to compensate only one impact (NRC, 2001). “All-in-One” trading model allows credit suppliers

to sell multiple kinds of credits generated in a single site, which will earn extra revenue and

lower the risk of selling only one type of credit.

Despite the advantages of stacking, only a few credit stacking programs are in operation or being proposed. There are issues that need to be carefully dealt with to create a credit stacking trading program. One challenge is to avoid double dipping. The program must make sure that the same credits, especially those overlapping two or more ecosystem functions, will not be sold more than once (Cooley and Olander, 2011). Another challenge is differentiating the additional credits that are generated only with the installation of conservation measures from

those that would occur naturally. These two issues are of critical importance because the

stacking program will otherwise result in a net loss of ecosystem service and will harm the

environment.

In water quality trading and other ecosystem services markets, a baseline is the key to

determining the amount of credits or offsets created by the use of conservation measures that

reduce pollutant emissions. Particularly when the ecosystem service credits are used to offset

pollution emissions reduction required to meet legal obligations, the selection of a baseline is

of critical importance. USEPA requires that the credits to be traded in water quality trading

must be “additional,” meaning that such reduction would not be made without the trading

program (USEPA, 2004). If credits were awarded to a reduction that would have occurred

anyway and sold to a regulated sector to substitute its required emissions reduction, the total

emissions reduction achieved might be lower than that achieved without trading, compromising the environmental objective associated with the regulation (Marshall and

Weinburg, 2012). Also, the selection of a baseline has significant implications for the costs of trading projects, including transaction costs, cost and quantity of credits, as well as the number and types of farmers who may participate and benefit from trading. A less stringent baseline would reduce the costs but might also add uncertainties in additionality. To balance costs and additionality, different trading projects define and use the baseline differently. For water quality trading, at least three ways exist to define a baseline in the United States:

1) When a baseline is usually used to describe the initial status prior to the time when an

action takes place, it could be called “time baseline.” This baseline could be the status of

emission immediately before the project starting day; it could also be the historical status

months or years before the project starting day. This is the less stringent type of baseline

compared to that in other definitions as it can be established empirically and used to easily

measure the changes from baseline.

2) In ecosystem services trading, the baseline is more often defined as the minimum control

level that sellers and buyers should meet separately before they can enter the trading

market and sell/buy credits. It can be called the “entry baseline”. Having an “entry baseline”

is a requirement from the USEPA (2007). A point source, usually the buyer in a water quality

trading project, needs to first meet a minimal on-site control requirement, the Technology-

Based Effluent Limit (TBEL). In order to receive the NPDES permit, a buyer can then

purchase credits to cover the difference between their TBEL and Water Quality-Based

Effluent Limit (WQBEL), a more stringent requirement that places limits on the discharges of

pollutants in the effluent, calculated to ensure water quality standards are met in the

receiving waterbodies. This baseline may become more important in the future in Ohio as

more stringent phosphorus limits is forecast for drainages to reduce the algal blooms in

Lake Erie, the Ohio River, and the Gulf of Mexico. For the sellers, the USEPA also defines a

baseline participation requirement (BPR), which is usually a minimum level of agricultural

management practices and/or compliance with existing federal, state and local laws and

regulation in the absence of trading. Only the reduction produced above and beyond this

level can be counted as credit for trading. This type of baseline is generally more stringent

than others and is used most frequently.

3) Some projects define the baseline as the performance goal that the projects aimed to

achieve. This baseline can be called the “goal baseline.” This type of baseline definition is

mostly used by regulatory agencies to describe the endpoint that exists after full

implementation of practices to meet Total Maximum Daily Load (TMDL) reduction

allocations. TMDLs exist for most of Ohio's watersheds and give precise loading limits for

NPDES permitted sites as well as goals for nonpoint source reductions. Projects using this

baseline may require the non-point source to meet TMDL load allocations or statewide

performance standards, or to install specific BMPs, or to reach a loading level even below

the set TMDL allocation.

The issue about baselines is a hot topic in water quality trading. In the 2015 EPA-USDA

National Workshop on Water Quality Markets, several participants from different sectors expressed the need for further clarification of baseline requirements for trading programs

(USDA and USEPA, 2016). In fact, some problems inherent in water quality trading baselines have been largely neglected but introduce significant uncertainties.

In all the baseline definitions discussed above, the pollution level on which credit calculations are based is usually measured in low flow scenarios, rather than in a high flow scenario—the flow of a stream during or after rain events. However, the majority of nutrient runoff occurs during high flow, especially during heavy rain events (Owens et al., 2010; Winslow,

2016), and is the main cause of nutrient loading to water bodies. Unfortunately, given the current technology, nutrient loading data during high flow are difficult to obtain although recently lower-priced automatic flow meters have improved. Since the nutrient loading that water quality trading aims to reduce happens mostly during high flow, using low flow measurements as the baseline, no matter how the baseline is defined or used, is problematic and introduces considerable uncertainties. Hydrological modeling is the main approach to

10 estimate high flow loading. However, using these models require detailed documentation of many aspects of each farm as well as local climate conditions. Without a good data set, the models do not yield reliable results; however, in most cases, building such a data set will increase the trading cost significantly.

The existing problems with current water quality regulations could also introduce uncertainties to a water quality trading baseline. For example, the Ohio EPA’s beneficial use designation for the protection of aquatic life classifies water bodies into several categories according to the existence of certain aquatic species. Although Ohio EPA has stated that

“beneficial use designations describe existing or potential uses of water bodies”, it is in reality only the existing uses of water bodies. The majority were classified as warm water habitats, with “typical assemblages of fish and invertebrates, similar to least impacted reference conditions”. This category is used as the baseline for most Ohio TMDL regulatory requirements.

The setting of this baseline therefore undermines the incentives to improve water quality since it is based on warmer conditions arising from post deforestation creating lack of canopy for streams, post creation of drainage tile lines, and tiling under most headwater springs. As an example, although Sugar Creek in Wayne and Holmes County of Ohio is fed by many springs

(cold water), it is classified as a warm water habitat by Ohio EPA, reflecting its existing, but not potential use. As a result, only minimum regulatory requirements are applied to Sugar Creek, while the opportunity to further improve the water quality is neglected. Similar issues can be found in the EPA anti-degradation policy. Therefore, it will be inevitable to review the current designated baseline and how it could affect the efficiency of water quality trading.

Another source of uncertainties comes from the heterogeneity of farms. Unlike point sources, the participating farms as a collective credit supplier have great heterogeneity among them. The potential and the cost of credit generation of each farm differs as the geological,

ecological, and management conditions are different. Setting a uniform baseline without taking

into account farm heterogeneity could potentially cause a loss of precision as well as of equity.

Some projects have attempted to address this issue; for example, the Southern Minnesota Beet

Sugar Cooperative Permit required farmers to provide a 5 year farm history. Then a load reduction baseline was set for each farm (USEPA, 2007). The Alpine Nutrient Trading Program in Ohio had dense "voluntary" sampling of approximately one sampling site per two square miles throughout the watershed, but few sites could accurately be aligned with the farms which implemented conservation measures (Richard Moore, personal communication).

1.4 Recent Advances of Water Quality Trading

In 2015, 22 operational water quality trading or offset projects were documented worldwide (Forest Trends Ecosystem Marketplace Initiative, 2016). Most of these projects were located in the U.S. (16 of the 22); others were found in Australia, New Zealand, and the U.K.

Collectively, these projects covered an area of 48,000 ha and transacted $31.1 million in credits, which was a significant increase compared with $20.8 million in 2013. A reduction of 30 million pounds of nutrients (mainly nitrogen and phosphorus) was achieved by these projects in 2015

(Forest Trends Ecosystem Marketplace Initiative, 2016).

In the U.S., some organizations are vigorously promoting WQT to improve the aquatic environment. For example, The Freshwater Trust has successfully operated multiple WQT

projects in Oregon. A thermal credit trading between the City of Medford’s wastewater

treatment plant and local landowners in the Rogue River watershed had led to the planting of

nearly 90,000 native plants along 25,109 feet of stream, reducing 594 pounds of nitrogen per

year and 438 million kilocalories from solar energy per day (The Freshwater Trust, 2016).

Implementation of this project saved the taxpayers about $8 million (The Freshwater Trust,

2016). Currently, this non-profit organization is actively exploring the opportunity to use WQT

to reduce nutrients in San Francisco Bay in California. Another example is the first multi-state

WQT project led by Electric Power Research Institute (EPRI) in the Ohio River Basin since 2012.

Beginning with Ohio, Indiana and Kentucky, this the world’s largest trading project aims to

cover eight states at its full scale, creating a market for 46 power plants, thousands of

wastewater utilities, and about 230,000 farmers (EPRI, 2014).

The World Resources Institute (WRI) is aiming to bring the WQT idea to other parts of the world. WRI has partnered with the decision-makers in Anhui, China, to use WQT for reducing nonpoint source nutrient pollution that caused eutrophication in Chao Lake. WRI has conducted studies for the feasibility of a point-nonpoint sources nutrient trading program in the Chao Lake watershed; has identified challenges in institution, policy, and capacity; and has developed recommendations to cultivate the water quality trading markets in China which can be coupled with government plans to create carbon markets as part of the Paris Climate Change

Conference of 2015 (Gao, 2016).

1.5 Motivation of Study

Most of the current scholarship on WQT focuses on identifying the obstacles in point- nonpoint source trading (Xie, 2014) such as the regulatory limitations (Abdalla et al., 2007; King,

2005), transaction cost (Fang et al., 2005), and risk and uncertainties (Ribaudo and Gottlieb,

2011; Walker et al., 2014) related to trading. However, limited knowledge about farmers’ participation in WQT as well as their adoption and selection of conservation measures in a WQT project is present (Breetz et al., 2005; Mariola, 2009). On the other hand, a good deal of scholarship exists on farmers’ adoption of BMPs. These studies examined the factors related to

BMPs’ adoption (Tosakana et al., 2010), evaluated the efficiency of programs such as the

Conservation Reserve Program (CRP) in promoting BMPs’ adoption (Wilson et al., 2013), and some (e.g., Morton et al., 2017) studied the impact of farmers’ experiences with climate change on their adoption of BMPs. However, WQT has not yet been included in those studies as a potential driver to enhance BMPs’ adoption rates.

This study aims to fill the knowledge gap between WQT and farmers’ selection and adoption of BMPs. In particular, I was interested in how trading models would affect the adoption of BMPs (Chapter 2) to inform the current debate over credit stacking with insights from the credit supplier side. I was also interested in expanding the use of WQT from agriculture to another often-neglected nonpoint source, the rural septic systems (Chapter 3).

Given that climate change is the context in which current and future water quality conservation efforts take place, I also studied the BMPs’ selection and adoption based on the farmers’ own observations of climate change and predictions of its impact on local water quality (Chapter 4).

A 14-page extensive questionnaire survey was developed to collect data for these studies

(Appendix I). The study focused on a subwatershed of Maumee River watershed, the Blanchard

River watershed, which eventually drains into western Lake Erie. The survey was conducted in

the rural area within the boundary of the Hancock County part of the Blanchard River

Watershed. The urban area and the area served with municipal sewer system were excluded.

Prior to constructing the survey, I interviewed a number of watershed and farming community

leaders such as the Farm Bureau and the Blanchard River Watershed Partnership who also

reviewed the survey first draft. There were two parts of the survey: Part One was about WQT

and BMP adoption and Part Two was about WQT and septic systems. The farming households

(those currently farming at least 80 acres of owned or rented farmlands) were asked to finish

both Parts One and Two of the survey, while the non-farming households completed only Part

Two. The “Drop-Off/Pick-Up” method (Melevin et al. 1999) was used to deliver the survey.

From October 2016 to February 2017, I and three local student assistants visited all eligible farming households (541 in total) and 359 randomly selected non-farming households in the

study area.

1.6 Objectives of Study

The general objective of this study is to evaluate whether WQT has the potential to serve as

an incentive to promote water quality conservation measures in the western Lake Erie basin.

Chapter 2 focused on the All-in-One credit stacking trading model and its potential to

increase farmers’ conservation measure adoption and willingness to participation in WQT. The

stacking approach has yet to be widely discussed in literature; limited discussions have focused

15 on the definitions, advantages, challenges, and solutions of stacking (e.g., Cooley and Olander

2011; Fox, 2008; Robertson et al., 2014), while few have studied the credit suppliers’ opinions of this idea. In this chapter, I studied farmers’ preferences for trading models, the All-in-One credit stacking model or the conventional single credit model. I also examined the correlation between the preference for the All-in-One trading model and the selection and adoption of

BMPs, testing the hypothesis that this model will lead to increased intention to adopt cover crops and no-till, which have the potential to generate multiple types of credits. In addition, I also tested whether this model will enhance farmers’ interest in participating in WQT. With a close examination of the social, economic, and behavioral factors potentially associated with farmers’ preferences for trading models and willingness to participate in WQT, this chapter also sheds light on improving the design of WQT projects that would be welcomed by the credit suppliers.

Chapter 3 explored the possibility of expanding the use of WQT from agriculture to rural septic systems, focusing on promoting septic system upgrades using WQT as a part of the community-based watershed management. As a nonpoint source of phosphorus, the contribution of septic systems to water quality degradation is often neglected with the presence of agriculture. In fact, domestic septic systems contributed as much as 88 tons/yr of

TP (total phosphorus) to Lake Eire (Ohio Department of Health, 2008). Septic system upgrade in northwestern Ohio, where the study site was located, is of special interest because the soil conditions there were limiting the effectiveness of pollutant removal. Following water quality trading programs that use a community-based approach such as the South Nation (O’Grady,

2011), Alpine Cheese and Muskingum plans (Moore, 2014 ), incorporating rural household

septic systems upgrade into water quality trading has the potential of broadening community

engagement within a county. It has the added benefit of bridging all rural residents--both

farmers and non-farmers—to address environmental and public health issues as a community.

In this chapter, factors associated residents’ interest in WQT for septic system upgrades were examined with correlation analysis; spatial analysis was also used to identify locations where households with the highest potential to participate in WQT were concentrated. Finally, I recommend the location and strategy for launching a WQT pilot project in the study area.

Chapter 4 investigates the impact of farmers’ observation of climate change on their adoption of conservation measures. Weber (2010) argued that personal observation and experience of climate change could be strong motivation for people to take action, while

Spencer (2011) found that the lack of first-hand experience of climate change’s adverse consequences is one reason that people do not take action. In an agriculture setting, Morton et al. (2017) found that some farmers perceived that there was too much uncertainty in climate change to justify the change of current agricultural practices and strategies; the possibility of having this perception was associated with their experiences with droughts, but not with floods.

Since farmers are close observers of climate, a more detailed study that examines their observation and experience of changes in the precipitation and temperature patterns is needed.

I also examined the relation between farmers’ observations of local climate change and their

selection and adoption of conservation measures, with a focus on cover crops and no-till. With

this study, I identified the key factors associated with farmers’ intentions to adopt additional

conservation measures under the climate change scenario.

References

Abdalla, C., Borisova, T., Parker, D., & Blunk, K. S. (2007). Water quality credit trading and

agriculture: Recognizing the challenges and policy issues ahead. Choices, 22(2), 117-124.

Arend, K. K., Beletsky, D., DePINTO, J. V., Ludsin, S. A., Roberts, J. J., Rucinski, D. K., Scavia, D.,

Schwab, D. J., & Höök, T. O. (2011). Seasonal and interannual effects of hypoxia on fish

habitat quality in central lake erie. Freshwater Biology, 56(2), 366-383.

Bosch, N. S., Evans, M. A., Scavia, D., & Allan, J. D. (2014). Interacting effects of climate change

and agricultural BMPs on nutrient runoff entering Lake Erie. Journal of Great Lakes

Research, 40(3), 581-589.

Breetz, H. L., Fisher-Vanden, K., Jacobs, H., & Schary, C. (2005). Trust and communication:

Mechanisms for increasing farmers’ participation in water quality trading. Land Economics,

81(2), 170-190.

Carroll, N., & Jenkins, M. (2013). The matrix: Mapping ecosystem service markets. Ecosystem

Marketplace. Retrieved from: http://www.ecosystemmarketplace.com/wp-

content/uploads/2015/09/the_matrix.pdf

Cooley, D., & Olander, L. (2011). Stacking ecosystem services payments: Risks and solutions.

Nicholas Institute for Environmental Policy Solutions, Working Paper NI WP, 11-04.

Cousino, L. K., Becker, R. H., & Zmijewski, K. A. (2015). Modeling the effects of climate change

on water, sediment, and nutrient yields from the maumee river watershed

doi:https://doi.org/10.1016/j.ejrh.2015.06.017

Dobiesz, N. E., Hecky, R. E., Johnson, T. B., Sarvala, J., Dettmers, J. M., Lehtiniemi, M., Rudstam,

L. G., Madenjian, C. P., & Witte, F. (2010). Metrics of ecosystem status for large aquatic

systems—a global comparison. Journal of Great Lakes Research, 36(1), 123-138.

Ecosystem Marketplace Initiative. (2016). Alliances for green infrastructure: state of watershed

investment 2016. Forest Trends. Retrieved from: http://www.forest-

trends.org/documents/files/doc_5463.pdf

Electric Power Research Institute (EPRI). (2014). Ohio River Basin Water Quality Trading Project.

Retrieved from: http://wqt.epri.com/pdf/3002001739_WQT-Program-Summary_2014-

03.pdf

Fang, F., Easter, K. W., & Brezonik, P. L. (2005). Point‐Nonpoint source water quality trading: A

case study in the minnesota river basin. JAWRA Journal of the American Water Resources

Association, 41(3), 645-657.

Fox, J. (2008). Getting two for one: Opportunities and challenges in credit stacking.

Conservation and Biodiversity Banking: A Guide to Setting Up and Running Biodiversity

Credit Trading Systems.

Gao, Y. (2016). China's response to climate change issues after paris climate change conference.

Advances in Climate Change Research, 7(4), 235-240.

Ho, J. C., & Michalak, A. M. (2017). Phytoplankton blooms in lake erie impacted by both long-

term and springtime phosphorus loading. Journal of Great Lakes Research, 43(3), 221-228.

Kalcic, M. M., Kirchhoff, C., Bosch, N., Muenich, R. L., Murray, M., Griffith Gardner, J., & Scavia,

D. (2016). Engaging stakeholders to define feasible and desirable agricultural conservation

in western lake erie watersheds. Environmental Science & Technology, 50(15), 8135-8145.

Kane, D. D., Conroy, J. D., Richards, R. P., Baker, D. B., & Culver, D. A. (2014). Re-eutrophication

of lake erie: Correlations between tributary nutrient loads and phytoplankton biomass.

Journal of Great Lakes Research, 40(3), 496-501.

King, D. M. (2005). Crunch time for water quality trading. Choices, 20(1), 71-75.

Levy, S. (2017). Learning to love the great black swamp. UNDark. Retrieved from:

https://undark.org/article/great-black-swamp-ohio-toledo/

Mariola, M. J. (2009). Are Markets the Solution to Water Pollution? A Sociological Investigation

of Water Quality Trading. The Ohio State University.

Marshall, E. P., & Weinberg, M. (2012). Baselines in Environmental Markets: Tradeoffs between

Cost and Additionality. United States Department of Agriculture, Economic Research

Service.

Melevin, P. T., Dillman, D. A., Baxter, R. K., & Lamiman, C. E. (1999). Personal delivery of mail

questionnaires for household surveys: A test of four retrieval methods. Journal of Applied

Sociology, 69-88.

Michalak, A. M., Anderson, E. J., Beletsky, D., Boland, S., Bosch, N. S., Bridgeman, T. B., Chaffin, J.

D., Cho, K., Confesor, R., Daloglu, I., Depinto, J. V., Evans, M. A., Fahnenstiel, G. L., He, L.,

Ho, J. C., Jenkins, L., Johengen, T. H., Kuo, K. C., Laporte, E., Liu, X., McWilliams, M. R.,

Moore, M. R., Posselt, D. J., Richards, R. P., Scavia, D., Steiner, A. L., Verhamme, E., Wright,

D. M., & Zagorski, M. A. (2013). Record-setting algal bloom in lake erie caused by

agricultural and meteorological trends consistent with expected future conditions.

Proceedings of the National Academy of Sciences of the United States of America, 110(16),

6448-6452.

Moore, R. H., (2014). The role of trading in achieving water quality objectives: Congress

testimony before the Water Resources and Environment Subcommittee Committee on

Transportation and Infrastructure United States House of Representatives. Washington D.C.

Morton, L. W., Roesch-McNally, G., & Wilke, A. (2017). Upper midwest farmer perceptions: Too

much uncertainty about impacts of climate change to justify changing current agricultural

practices. Journal of Soil and Water Conservation, 72(3), 215-225.

O'Grady, D. (2011). Sociopolitical conditions for successful water quality trading in the south

nation river watershed, ontario, canada. Journal of the American Water Resources

Association, 47(1), 39-51.

Ohio Department of Agriculture, Ohio Department of Natural Resources, Ohio Environmental

Protection Agency & Ohio Lake Erie Commission. (2013). Ohio Lake Erie phosphorus task

force II final report. Retrieved from:

http://lakeerie.ohio.gov/Portals/0/Reports/Task_Force_Report_October_2013.pdf

Ohio Department of Health (ODH). (2008). ODH report to the Ohio Lake Erie phosphorous task

force. Retrieved from:

http://www.epa.ohio.gov/portals/35/lakeerie/ptaskforce/041808ODHoleptfRPT.pdf

Ohio Environmental Protection Agency (Ohio EPA). (2010). Ohio Lake Erie phosphorus task

force final report. Retrieved from:

http://www.epa.state.il.us/water/nutrient/presentations/lake_erie_task_force.pdf

Owens, L., Bonta, J., & Shipitalo, M. (2010). USDA-ARS north appalachian experimental

watershed: 70-year hydrologic, soil erosion, and water quality database. Soil Science

Society of America Journal, 74(2), 619.

Ribaudo, M. O., & Gottlieb, J. (2011). Point-nonpoint trading - can it work? Journal of the

American Water Resources Association, 47(1), 5-14.

Robertson, M., BenDor, T. K., Lave, R., Riggsbee, A., Ruhl, J., & Doyle, M. (2014). Stacking

ecosystem services. Frontiers in Ecology and the Environment, 12(3), 186-193.

Scavia, D., Allan, J. D., Arend, K. K., Bartell, S., Beletsky, D., Bosch, N. S., Brandt, S. B., Briland, R.

D., Daloğlu, I., & DePinto, J. V. (2014). Assessing and addressing the re-eutrophication of

lake erie: Central basin hypoxia. Journal of Great Lakes Research, 40(2), 226-246.

Selman, Mindy, Evan Branosky and Cy Jones. 2009. Water quality trading programs: An

international overview. World Resources Institute. Retrieved from:

http://www.wri.org/publication/water-quality-trading-programs-international-overview

Stumpf, R. P., Wynne, T. T., Baker, D. B., & Fahnenstiel, G. L. (2012). Interannual variability of

cyanobacterial blooms in lake erie. PloS One, 7(8), e42444.

The Freshwater Trust. (2016). Uplift report. The Freshwater Trust. Retrieved from:

https://www.thefreshwatertrust.org/wp-content/uploads/2017/10/TFT-Uplift-Report-

2016-web.pdf

Tilman, D., Fargione, J., Wolff, B., D'Antonio, C., Dobson, A., Howarth, R., Schindler, D.,

Schlesinger, W. H., Simberloff, D., & Swackhamer, D. (2001). Forecasting agriculturally

driven global environmental change. Science (New York, N.Y.), 292(5515), 281-284.

Tosakana, N., Van Tassell, L., Wulfhorst, J., Boll, J., Mahler, R., Brooks, E., & Kane, S. (2010).

Determinants of the adoption of conservation practices by farmers in the northwest wheat

and range region. Journal of Soil and Water Conservation, 65(6), 404-412.

USDA Office of Environmental Markets & USEPA Office of Water. (2016). Reports of 2015 EPA-

USDA national workshop on water quality markets. Retrieved from:

https://www.epa.gov/sites/production/files/2016-

07/documents/cleared_epa_usda_workshop_report.pdf

U.S. Environmental Protection Agency (USEPA). (2004). Water quality trading assessment

handbook: Can water quality trading advance your watershed’s goals? Washington D.C.

U.S. Environmental Protection Agency (USEPA). 2007. Water quality trading toolkit for permit

writers. Washington D.C.

Wainger, L. A. (2012). Opportunities for reducing total maximum daily load (TMDL) compliance

costs: Lessons from the chesapeake bay. Environmental Science & Technology, 46(17),

9256-9265.

Walker, S., & Selman, M. (2014). Addressing risk and uncertainty in water quality trading

markets. World Resources Institute,

Walsh, J., Wuebbles, D., Hayhoe, K., Kossin, J., Kunkel, K., Stephens, G., Thorne, P., Vose, R.,

Wehner, M., Willis, J., Anderson, D., Doney, S., Feely, R., Hennon, P., Kharin, V., Knutson, T.,

Landerer, F., Lenton, T., Kennedy, J., and Somerville, R. (2014). Our changing climate:

Climate change impacts in the United States. The Third National Climate Assessment.

Washington, D.C.

Willamette Partnership, World Resources Institute & National Network on Water Quality

Trading. (2015). Building a water quality trading program: Options and considerations.

Willamette Partnership. Retrieved from: http://willamettepartnership.org/wp-

content/uploads/2015/06/BuildingaWQTProgram-NNWQT.pdf.

Wilson, R., Burnett, L., Ritter, T., Roe, B., & Howard, G. (2013). Farmers, phosphorus and water

quality: A descriptive report of beliefs, attitudes and practices in the Maumee watershed of

northwest Ohio. The Ohio State University, School of Environment & Natural Resources.

Winslow, C.J. (2016). Current Research Efforts on Nutrient Load Reduction Methods. Region 5

Harmful Algal Bloom Clean Water Act and Safe Drinking Water Act Workshop.

Xie, Y. (2014). Watershed modeling, farm tenancy and adoption of conservation measures to

facilitate water quality trading in the Upper Scioto watershed, Ohio. The Ohio State

University.

Chapter 2. “All-in-One” Credit Stacking Water Quality Trading Model and BMP Adoption in

the Blanchard River Watershed of Ohio

2.1 Introduction

This chapter aims to test whether the concept of stacking ecosystem service credits (termed

“All-in-One” water quality trading model in this study) is favored by the farming community in the Blanchard Watershed of Ohio, USA, and whether this model will lead to increased intention to adopt no-till plowing and cover crops and farmers’ interest in WQT program. Defined by

Daily (1997), ecosystem services are “the conditions and processes through which ecosystems, and the species that make them up, sustain and fulfill human life.” It is a human-centered concept in which nature is the provider of benefits that could be used by humans. There are four categories of ecosystem services (Millennium Ecosystem Assessment, 2005): 1) provisioning ecosystem services that provide products such as fresh water and fish, etc.; 2) regulating ecosystem services that regulate the critical processes related to human lives, such as pollination and water purification; 3)supporting ecosystem services that maintain the living conditions for humans, such as the provision of habitat and formation of soil; and 4) cultural ecosystem services that support the cultural, aesthetic, religious, and social needs of humans, such as tourism and recreation. Based on the concept of nature’s benefits, different approaches were developed to measure the value of ecosystem services. Usually an estimated price is given to a certain ecosystem based on the services it provides. For example, Costanza et al. (1998) estimated the value of the ecosystem services provided by the entire biosphere to be 16 to 54

trillion USD per year. Updated estimates of the value were $145 trillion/yr in 2007 and $125

trillion/yr in 2011 (Costanza et al., 2014). The latter number which decreased in value was due

to land use changes.

Meanwhile, the purchase and sale of ecosystem services and the idea of creating a market

for ecosystem services were gradually developed since Costanza's (1998) pioneering article. For example, Carroll and Jenkins (2013) provided a matrix funded by the Betty and Gordon Moore

Foundation titled "Innovative markets and market-like instruments for trading ecosystem services." The main categories of this growing matrix were "compliance forest carbon, voluntary forest carbon, Redd fund-based carbon financing, compliance water quality trading, voluntary private sector watershed payments, payments for watershed services and water funds, environmental water rights purchases, compliance biodiversity compensation, voluntary biodiversity compensation, government-mediated biodiversity payment for ecosystem services, recreation (ecotourism, park fees, hunting licenses), genetic resources (access and benefit sharing), marine resource markets, certified agricultural products, and certified forest products”

(Carroll and Jenkins, 2013 ). Of these, carbon financing and certified forest products top the list for high value, mainly owing to the issues of carbon's leading role contributing to climate change.

Many organizations and individuals involved in environmental issues considered the ecosystem services market promising for solving important environmental problems, both practical and theoretical (van Maasakkers, 2016). The advantages of ecosystem services markets include 1) reducing conservation cost, for example, it might be cheaper for a city to pay

27 for the upstream conservation of its drinking water source watershed than to pay for additional mechanical filtration; 2) promoting technology renovation—those that use new technologies that produce less pollution will be able to sell the pollution reduction as credits and be more competitive; 3) provide incentive for landowners, especially those not being regulated by current legislation, to adopt conservation measures, since they will earn extra revenue by selling the credits generated by the conservation measures. Water quality trading (WQT) is a form of ecosystem service markets. It is a compliance approach that allows point sources, such as factories and wastewater treatment plants, to meet their regulatory obligations (e.g.,

National Pollutant Discharge Elimination System (NPDES) Permit) by using pollutant reductions created by another source, such as agriculture, that has lower pollution control costs (USEPA,

2004).

The estimation of the monetary value of ecosystem and its services, as well as the creation of market for these services, has been mostly studied through the lens of free-market neoclassical economics. When it comes to water quality trading, where farmers are the main suppliers of nutrient reduction services, these studies often assumed that farmers are motivated to participate in the trading activity because they are rational, self-interested agents whose goal is to maximize their own benefit (Popkin, 1979). However, others argued that, comparing with the rationality of maximizing economic benefits, farmers give priority to other values, which are not always aligned with rational choices. Anthropologists and economists call this the “substantivist approach”, first coined by Karl Polanyi in his work The Great

Transformation (1944). The term “substantivism” to Polayni refers to the broad sense of provisioning as opposed to economics is the way society meets material needs. Polanyi

asserted that pre-market societies use substantivist economics. Anthropologists have emphasized cultural morals, values, kinship, social relations, politics and religion, embeddedness and nonmarket factors in the substantivist approach and hold that these ideas are used alongside of capitalist ideas and contrast this with a pure “rational choice” approach.

The key journal for sustantivist approaches is Research in Economic Anthropology. My dissertation results support this substantivist approach.

Political historian James Scott (1976) argued that “the subsistence ethic”, or the survival

and stable life, is the top priority of farmer in Burma and Vietnam. The rational choice theory is also criticized by Green and Shapiro (1996) to be methodologically defective with limited sound empirical support. In fact, empirical studies of different fields, such as political science, behavioral science and rural sociology, have repeatedly found evidences against the rational choice theory. Moreover, the reductionist point of view of the rationality theory is also problematic, especially when it comes to complex issues such as nature-human interaction.

WQT as an effort to restore the function of the agricultural ecosystem, emphasizes the emergent properties of ecosystems from integrating human values, beliefs, social capital and institutions, as well as the ecological structure and function of agroecosystems (Moore, 2009).

Therefore, to conceptualize WQT as a free-market activity that involves only rational participants might lead to failure.

In practice, although the economic benefits seem tempting, farmers, who are suppliers of water quality credits, have not been participating as much as expected (King and Kuch, 2003;

Kramer, 2003). Despite offering financial incentives and technical assistance, nearly every water

quality trading program that involves farmers has experienced difficulties convincing farmers to

participate (Breetz et al., 2005). Many studies have tried to understand farmers' reluctance to

voluntarily participate in water quality trading programs. One problem is that participating in

water quality trading could introduce greater uncertainties such as loss of productivity and

profitability to farmers by using conservation measures (McSweeny and Kramer, 1986) that

require the opportunity costs (money, time and land) of using new practices. In addition, risks

such as loss of farm operation autonomy, increased government oversight, and negative

publicity about farmers as polluters (Breetz, 2005) also discourage farmers from participating in

water quality trading. Since the farming business is already subject to the variable weather and

markets, farmers are naturally reluctant about introducing more uncertainties and risk from

participating in WQT. Psychological factors are also important to farmers’ willingness to

participate in WQT. For example, Edward-Jones et al. found that openness to innovation could

explain 20%-60% of the variation of environmentally-oriented behavior. Social barriers, such as

communication mechanisms and trust in the program administrator are important factors but are often neglected in the literature (Breetz, 2005). Mariola (2009) argued that farmers behaved simply as rational economic actors and the idea that the supply of water quality credits only followed demand was a flawed assumption, as it dismissed social factors. He found that the social embeddedness of economic actors or the social relationships between both economic and non-economic actors (Hess 2004) and the trust relations among them is a critical factor for the success of a water quality trading programs. Methorst et al. (2017) have argued that

"In decision-making on farm strategies, the family farm is inherently intertwined with

pre-existing socio-material structures; the farmer does not and cannot make strategic

decisions as on a blank canvas. The existing sociomaterial context both enables and restricts

farm development, as it offers opportunities as well as limitations. In other words, there is

‘room for manoeuvre’ to act within the socio-material context."

Equally, this room for maneuvering in farm decision-making takes place in the grey area

between private property rights and use rights while ecosystem services tend to redefine

commons (use rights) in the process of commoditizing them (Richard Moore, personal

communication). These factors, in addition to the scientific and institutional soundness of the

WQT programs, need to be addressed in the design process of any water quality or ecosystem

service trading program.

The term “All-in-One” trading model in this chapter refers to the concept of “stacking”.

“Stacking” is a recent attempt that aims to make water quality trading more economically feasible. “Credit stacking” is defined as “selling credits representing two or more spatially overlapping ecosystem services as separate commodities, each compensating for different permitted impacts” (Robertson et al., 2014). Since different ecosystem services can spatially overlap, the credits representing different types of ecosystem services could also be separately stacked. Different types of credits could be sold separately to compensate different environmental impacts (Robertson et al., 2014). Although only a few credit operational or proposed stacking programs exist, the situations that stacking applies to vary. Robertson et al.

(2014) summarized that stacking has been applied 1) to situations where a single spatial unit as

a whole can generate multiple credit types, which can be sold separately (Willamette

Partnership, 2010); 2) within a spatial unit where different credit types are generated in

different subunits that are not spatially overlapping or lineated (NCEEP, 2009); or 3) where

multiple ecosystem services generated within one property are bundled and sold to

compensate only one impact (NRC, 2001).

Stacking is considered a promising conservation tool by many regulators and practitioners

who want to promote ecosystem service markets, but its use remains controversial. Regulators

believe that stacking creates incentives for landowners to add additional conservation

measures to increase their ecological value and to produce more credits. Stacking allows credit

providers to sell multiple kinds of credits generated in a single site, which will earn extra

revenue and lower the risk of selling only one type of credit so that if the particular

conservation measure failed to produce one type of credit, it still might produce other types of

credits. For example, planting trees might sequester carbon and lower stream temperature but

be less inefficient at holding soil runoff related to decreasing phosphorus and nitrogen. There

will also be seasonal differences in these varying types of credits making different ones

vulnerable depending on the seasonal weather. More importantly, stacking may also improve

the design of conservation projects in order to produce multiple ecosystem services at one site.

Stacking may also be useful for financing the best conservation practice when that practice is

ecologically sound but cost prohibitive. Ecosystem services can be based on either or both

ecosystem structure or function as defined by Eugene, P. Odum (1962). Simple trading might

be based on either phosphorus, nitrogen, or carbon. Taken by themselves, these would be

Odum's "structure" as they are singular building blocks of an ecosystem. However, each of

these has cycles, such as the carbon cycle, which Odum classified as the process necessary for

ecosystem "function". Compared to conventional single credit trading, stacking has the potential to incorporate more fully both ecosystem structure and function leading to biocomplexity.

Despite the above advantages, there are issues that need to be carefully dealt with to create a credit stacking trading program. The first challenge is to guarantee “additionality.” The ecosystem service credit must be additional, such that the ecosystem service would not occur without the conservation measure (Cooley and Olander, 2011). The second challenge is to avoid

double dipping. The trading program must guarantee that the same ecosystem service credit,

especially those that overlap two or more ecosystem functions, will not be sold more than once

(Cooley and Olander, 2011). If these two issues are not well resolved, a net loss of ecosystem

service and harm to the environment will result. Another challenge is that stacking could only

arise where multiple types of ecosystem functions are regulated by the government, since

currently stacking is only allowed in the regulatory market (buyers buy credits in order to meet

regulatory requirements), not in the voluntary market (Robertson et al., 2014). However, a

third party could voluntarily buy a part of the stacked credits for environmental

goodwill/sustainability programs. Thus, the categories listed in the Ecosystem Marketplace

matrix mentioned previously can potentially be linked. The new funds generated through

linking categories could provide additional income to the trading program.

Stacking as an innovative approach has yet to be well described in the literature and is limited to a few sources (e.g., Cooley and Olander, 2011; Fox 2008; Robertson et al., 2014).

Most of their discussions have focused on the definitions, advantages, challenges, and solutions

of stacking, while few have studied the perceptions and opinions of ecosystem service trading

participants to stacking. In this study, we aimed to 1) test the hypothesis that farmers prefer

stacking over single credit trading, 2) test the hypothesis that stacking will lead to adoption of

conservation measures that could provide multiple ecosystem services; and 3) test the

hypothesis that stacking will result in increased willingness for farmers to participate in water

quality trading. In addition, we also investigated the social, economic, and behavioral factors

potentially associated with farmers’ preferences of trading models and willingness to

participate in water quality trading.

2.2 Methodology

2.2.1 Study Area

Agriculture in the Maumee River Watershed was identified as the main contributor to the excessive nutrients in western Lake Erie basin. The Blanchard River Watershed in Northwest

Ohio is one of the subwatersheds of Maumee River Watershed, covering 493,415 acres across five Ohio counties: Hancock, Hardin, Putnam, Allen and Wyandotte. Over 80% of Blanchard

River Watershed is cropland. This study focused on the part of the Blanchard River watershed that falls within Hancock County. Hancock County takes up about 50% of the area in the watershed with nearly 3/4 of the county being in the Blanchard River watershed. About 68% of the land in Hancock County is farmland, of which 92.4% is cropland. Corn and soybeans are the principal crops and Hancock County lies in the eastern edge of the US Corn Belt. Eighty-three

percent of the cropland is managed by 250 farms of 260 acres or more each. The remaining 17%

of the cropland is managed by 443 farms managing 259 acres or less each. Agriculture is the

leading industry of Hancock County and in 2012 the total crop sales were $146,881,000 and

livestock sales were $13,364,000 (USDA, 2012).

According to the US Census Bureau (2017), Hancock County had a population of 75,573 and

31,083 households at the end of 2015. The city of Findlay, located in the low land of the former

Black Swamp, had a population of 41,202 and is the county seat.

2.2.2 Survey

A questionnaire survey was conducted within the boundary of the Hancock County part of the Blanchard River watershed. Eligible households were those currently farming at least 80

acres of owned or rented farmland in the study area. According to the Hancock County Auditor

and 2012 Census of Agriculture (USDA, 2012), 541 households met this criterion.

The “Drop-Off/Pick-Up” method (Melevin et al., 1999) was used for this survey. This method

has the advantage of reducing the nonresponse bias by increasing response rate (Steele et al.,

2001), especially in natural resource surveys (Allred et al., 2010), as the survey could be

explained to the potential respondents in person, and the difficulty of returning the survey by

mail is avoided. The face-to-face communication also allows the researcher to better determine

the eligibility of the respondent and gain experiential insights (Allred et al., 2010; Steele et al.,

2009). The personal contact also allows the research to collect qualitative information, such as

participant observation, discussion on related topics as well as site visit of farms. In this study,

the addresses of these households were obtained from the Hancock County Auditor for survey

delivery. Three local student assistants who helped administer the survey and I were certified

by the Collaborative Institutional Training Initiative (CITI) for human subjects and the survey

was approved by the Ohio State University Institutional Review Board of the Office of

Responsible Research.

From October 2016 to February 2017, we visited all 541 eligible households in the study

area in person. In 392 households there was no adult present at the first and second time we

visited, so we were not able to use these households. Of the remaining 149 households, 96

completed the survey and 53 refused to complete the survey (Table 2.1). The response rate was

64.4%.

Farming Household

Completed 96

Refused 53

No adult available 392

Total 541

Response rate 64.4%

Table 2.1 Responses of farmers’ survey.

The survey questionnaire included 5 sections. The first section focused on farmers current farming practice and concerns about local water quality issues; the second section studied their perceptions of WQT models and their choice of BMPs in the WQT scenario; the third section studied their observation of climate change and their choice of BMPs in the climate change scenario (this result of this section will be discussed in Chapter 4); the fourth section was about the characteristics of the farm and the household; and the last section was about their household septic systems (will be discussed in Chapter 3). The BMPs listed for farmers to

choose were the same for the WQT and the climate change scenarios. The listed BMPs were those being actively promoted in the area and/or have been studied in other studies. The two- stage ditch conservation practice was excluded because it was relatively new in the study area and remained controversial during the time of study. The 4R nutrient stewardship was not included as a BMP, rather, the concept of improving fertilizer efficiency which is a key component of the 4R approach was reflected in the listed BMPs such as improving soil quality and using precision agricultural technology. Tree planting along streams were removed from the list after field testing of the questionnaire, because the trees and logs along streams were considered the cause of flooding and other problems in the area, having such a BMP was likely to trigger resentment and reduce response rate.

2.2.3 Analytical Approach

A binary logistic regression was employed to analyze the relationship between farmers’ interest in WQT (Yw) and their preference for the trading model, as well as other variables,

including the concerns over water quality issues, current conservation measures, demographic characteristics of farmers and characteristics of farms (Table 2.2).

Although the Drop-off/Pick-up survey method reduces nonresponse bias by lowering the

nonresponse rate, nonresponse bias still needs to be addressed. Following the “comparison of

early to late respondents” method recommended by Lindner et al. (2001), I defined respondents who returned their completed survey in the first pick-up attempt to be early

respondents, while those who did not complete the survey in the first pick-up but returned in

37 the second or later attempts to be late respondents. The two groups were compared in interest in WQT (YW), preference to trading model, concerns over water quality issues, current adoption of conservation measures, characteristics of the farmers and farms, and knowledge about WQT, no significant difference were found. This also corresponds with my own observation from interviewing locals.

Variables Label Description Coding Dependent Variables Interest in WQT YW Binary 0: No 1: Yes

Independent Variables Preference to the “All-in-One” trading AllinOne Binary 0: No model 1: Yes Concerns over water quality issues Total Concern Score T.Concern Numeric Combination of concern ratings of all water quality issues listed in the survey (range: 0-39) Current adoption of conservation measures Tillage TillRank 4-class 1: Conventional tillage ordinal only 2: Conventional and conservation tillage 3: Conservation tillage and no-till 4: No-till only Cover crops CoverCrop Binary 0: No 1: Yes Total number of BMPs currently in use TotalUsing Numeric Combination of adopted conservation measures listed in the survey (range: 0-15) Characteristics of the Farmers Age Age Numeric Gender Gender Binary 0: Female 1: Male Education Education 4-class 1: High School ordinal 2: Some college 3: College graduate 4: Graduate degree Years lived in Blanchard River watershed Live_yr Numeric Table 2.2 Variables for the binary logistic regression.

Variables Label Description Coding Characteristics of the Farms Land owned Own Numeric Land leased Lease Numeric Total farmland (owned + leased) Land Numeric Percentage of leased land in total LeaseP Numeric farmland Years managing the farm Manage_yr Numeric Farming income Income 5-class 1: Less than $24,999 ordinal 2: $25,000 - $49,999 3: $50,000 - $74,999 4: $75,000 - $99,999 5: More than $100,000 Role of farming income to a household Income_role 4-class 1: < 10% ordinal 2: 10-50% 3: 51-90% 4: > 91% income Knowledge about WQT Knowledge Binary 0: No 1: Yes Table 2.2 Continued.

2.2.4 Limitations of Methodology

Given the limitation of methodology as well as time and resources, there is some bias of the

survey. First, although the face-to-face interaction of the Drop-off/Pick-up method could

increase response rate, it also significantly increased the time and cost of conducting the survey,

especially when the households were less assessable. In this survey, many households had no

one home when the questionnaire was delivered. The missed households (non responders)

might introduce bias and I did not go back and take the extra time to find them and interview

them. We made sure at least one drop-off attempt was made to every potentially eligible

household within the study area. Second, the personal interaction could introduce bias to this

study because the personality of the researcher and the way he/her approached the potential respondents might have affected the response rate as well as the answers to the questions.

Three local college students were hired to deliver the survey for a short period of time at the

beginning of the research. The assistants contributed less than 5% of the total responses, while

the rest was conducted by the author alone. The bias from assistants should be minimum.

However, the interaction between the author and different potential respondents varies in every approach. To minimize the personal impact, the author used consistent language to explain the survey and kept the style of approaching consistent. Third, another source of bias was that, it can be difficult to control whether the farmer actually completed the questionnaire, and/or answered the questions honestly. This bias was treated with triangulation in this study.

To the most important questions, such as farmers’ choice of BMPs, the same list of BMPs was provided in two different scenarios, WQT and climate change. The results (ranking of BMP popularities and factors significantly associated with the BMP choice) were very similar in both scenarios. Also, during the survey, many farmers expressed the desire of communicating their opinions to people outside the local farming community via this survey. The response from these farmers might be considered credible. Other qualitative data collected from communication with farmers during the survey could also be used to support the findings of the questionnaire. Although the uncertainties regarding veracity could not be completely eliminated, the biases should be within a reasonable range given the triangulation methods discussed. For BMP adoption, it is difficult to improve veracity by cross-checking, since the preference for BMP and the factors associated with it varies from study to study, even within similar regions (Knowler and Bradshaw, 2007; Prokopy et al., 2008; Liu et al., 2018). This is also a limitation of this study.

2.3 Results

2.3.1 Characteristics of Farmers and Farms

All farmers who completed the survey were crop growers; 34.4% of them also kept livestock.

The average age of farmers was 60.78 years, ranging from 25 to 85 years. The majority was between 50 and 80 years old (Figure 2.1). The distribution of farmers’ age was similar to that of the national average (USDA, 2012). All farmers who responded were high school graduates or higher, 26.7% of them had a college degree and 6.7% had a graduate degree. Of the farmers who responded, 85.6% were male. Of the farmers' families, 81.7% had lived in the Blanchard

River watershed for over 50 years and 35.5% for over 100 years. Most farmers (87.1%) were the heads of their households; 45.0% of the farmers made farming decisions by themselves alone;

21.5% made decisions with their partners; and 20.4%, with their spouses.

Figure 2.1 Distribution of farmers' age (in years).

The land managed by the farmers who responded totaled 56,905 acres, taking up 24.7% of the total farmland in Hancock County, Ohio. Of this farmland 49.1% was owned by the farmers, and 50.9% was leased. Of the farming households, 59.1% had been managing their family farms for over 50 years and 22.6%, for over 100 years. For 33.3% of the farmers, their farming income was between $25,000 and $50,000 per year; for 25.8%, it was over $100,000 per year (Figure

2.2). For the majority of farmers, farming was not the only income source of their families; only

26.7% of the farmers relied mostly on farming (Figure 2.3).

Figure 2.2 Distribution of Household Farming Income (in US dollar).

Figure 2.3 Role of Farming in Total Household Income.

2.3.2. Current Adoption of BMPs

The survey found that 97.7% of the farmers used the nearby streams as an outlet for drainage of their farm fields; 22.7% used the streams for entertainment purposes, such as fishing, hunting, hiking, etc.; and only 5.7% used the streams as a source of drinking water for

livestock.

The adoption rate of no-till and conservation tillage was high in the study area: 76.34% of

the farmers have adopted no-till while 58.06% have adopted conservation tillage; only 35.48%

still used conventional tillage (Figure 2.4). In fact, most of the farmers used more than one type

of tillage practice. Only 8.6% used only conventional tillage while over 90% have been practicing

43 either no-till or conservation tillage or both. The no-till only farmers comprised about 14% of all the responses (Figure 2.5).

Figure 2.4 Tillage Currently In-use.

Figure 2.5 Most farmers used more than one type of tillage.

The adoption rate of other BMPs was also high. The most popular BMP was soil testing— about 78% of the farmers were using soil testing to improve fertilizer application precision.

Cover crops were also welcomed by the farmers; about 63% of them had grown cover crops on at least some parts of their farms. The adoption rates of cropping down (the non-application of phosphorus fertilizer to decrease soil phosphorus), reducing herbicides and pesticides use, grass waterways, improving drainage, and improving soil quality were nearly 60% (Table 2.3).

Frequent crop rotation and buffer strips were less popular, but still about 50% of the farmers had adopted these BMPs on their farms. Since only 34.4% of the farmers kept livestock, the adoption rate of manure management BMPs was low.

BMPs Adoption Rate

No-till or conservation tillage 90%

Using precision ag technology (e.g. soil test) to vary fertilizer application rates 78% within fields

Planting cover crops 63%

Cropping down (example: not applying phosphorus fertilizer on a crop that uses 63% it because the soil may already have extra phosphorus)

Reducing application of herbicides or pesticides 62%

Installing grass waterways 59%

Installing or improving tile drainage (e.g. add new tile, end of ditch structure) 58%

Improving my soil quality (includes increasing humus or adding soil amendments, 57% etc.)

Rotating other crops (small grains, forages, others) more frequently in corn-soy 49% rotations

Installing buffer strips/ filter strips 48%

Avoiding manure spreading on frozen ground 27% Table 2.3 Adoption rate of BMPs.

BMPs Adoption Rate

Avoiding manure spreading near waterways 26%

Used nutrient management plan to decide manure application rates 22%

Fencing livestock from the stream/ditch 13%

Table 2.3 Continued.

2.3.3 Farmers’ Preference for WQT Models: “All-in-One” Credit Stacking Model vs. Single

Credit Model

The term “All-in-One” trading model is a form of credit stacking WQT where the credits are

summed for the benefit of the seller of credit. The use of All-in-One model in the survey was for

easier communication of the credit stacking concept to farmers. This study hypothesized that farmers would have a preference for the All-in-One model over the conventional trading model

that allows only one type of credit, such as nitrogen or phosphorus, to be traded in one

program. Since most farmers were not familiar with even the basics of WQT, a written

introduction of the concepts of the All-in-One model and single credit trading model were

presented in the survey, and the farmers were asked which model they preferred after reading

the introduction. The survey found that 70.8% of the respondents preferred the All-in-One

model, 16.7% preferred the more commonly seen phosphorus trading, and 12.5% preferred

trading nitrogen or temperature or biodiversity credits alone (Figure 2.6).

Figure 2.6 Farmers’ Preference for Trading Model.

2.3.4. Would All-in-One Trading Model Encourage Cover Crops and No-Till Adoption?

We hypothesized that farmers’ preference for the All-in-One trading model would lead to increased adoption of additional cover crops and no-till. In the survey, farmers were asked which BMPs they would add to their farms based on the trading model they chose. We found that, for the farmers who preferred All-in-One trading model, planting cover crops was the most popular BMP, followed by improving drainage, reducing herbicide and pesticide use, and doing soil testing; no-till, on the other hand, received fewer votes than other BMPs (Figure 2.7).

However, the logistic regression results showed that the preference for trading model was not significantly related to farmers’ willingness to add cover crops or no-till (Table 2.4). This finding

implies that, trading models per se were not a determinant of farmers’ decision of cover crops

or no-till adoption.

In fact, farmers’ decision making process of BMP adoption is very complicated. Many factors

have been identified to be related to farmers’ adoption of BMPs by various studies. However, most of these factors were found inconsistent (Knowler and Bradshaw, 2007; Prokopy et al.,

2008; and Liu et al., 2018). In this study, I found that the adoption of cover crops was only related to its current usage, farmers who had already adopted cover crops were more likely to continue or expand the use of cover crops. As other variables were held constant, the odds of a farmer who had adopted cover crops to adopt additional cover crops was 6.03 times higher than those who had not already adopted. Zhong et al. (2015)’s study in Kentucky found that previous experiences would significantly increase the willingness of farmers to expand the use of BMPs through WQT programs. But they also found that this pattern did not apply to the adoption of no-till. Similarly, in this study, the willingness to adopt additional no-till was not influenced by the current usage of tillage types, rather, it was significantly related to total household annual income and the area of leased land. The lower the household income and/or the larger the percentage of leased land to owned land, the less likely farmers would adopt no-

till. Zhong et al. (2015) also reported that farmers who rent more land were less willing to adopt no-till; while the cost savings feature of WQT programs was an effective factor to positively motivate the use of several BMPs, it failed to motivate no-till adoption. Oppositely, studies such as Movafaghi et al. (2013) and Xie (2014) found that farmers thought that no-till had more

economic benefit than environmental benefit so it met the needs of renters who looked for

short term economic return. Based on this assumption, farmers with higher leased land/total

48 land ratio and higher farming income/total income ratio were more likely to adopt no-till. But it was not the case in this study. This study and Xie’s study both took place in Ohio and Zhong et al.’s study took place in the neighboring state Kentucky, the contradicting results imply that the influence of economic benefit on no-till adoption varies even within similar regions such as the

Ohio River Valley.

This study found different determining factors for no-till and cover crops adoption. The difference might be related to the characteristics of the two BMPs, as well as their different stages of adoption in the study area. The process of BMP adoption is a temporally continuous, dynamic learning process consisting of several stages. First, farmers have to be aware of the availability of a BMP; second, farmers become interested in the BMP and try to learn more about it and consider its applicability on their own farms; third, farmers test and evaluate the performance of the BMP before widespread adoption in order to reduce risk and develop skills; the last stage is to scale up the adoption and customize it to fit the conditions of their farms

(Pannell and Vanclay, 2011; Rogers, 2010; Pannell, 1999). In the Blanchard River watershed, many farmers were in the experiment stage of cover crops adoption. The result was satisfying enough that farmers would consider continuing and/or expanding. On the other hand, no-till and conservational tillage had been adopted for a longer time and for a wider range with over

90% of farmers having had experience using no-till and conservational tillage. However, only 14% of the farmers used no-till exclusively. There might be some barriers, such as the cost and concern of using herbicides, preventing farmers moving from conservational tillage to exclusive no-till. Moreover, one farmer who started no-till in the 1970s indicated that, in the past year his no-till field had much less yield than conservational and conventional tillage fields, making

him wanting to give up no-till. These kinds of problems could not be easily solved by WQT models.

This study also investigated farmers’ concern regarding BMP adoption (Figure 2.8). The

issues that concerned most farmers were how the yield would be affected by adopting BMPs.

As noted by Shaffer and Thompson (2013) and Brandt and Baird (2008), when adopting water quality BMPs, the major barrier was farmers’ concerns of yield and income risks. The second most concerned issue was how much improvement in water quality the BMPs could make.

Concern over BMP efficacy was also the issue determining farmer participation in WQT (Figure

2.9). These findings reflected the environmental awareness and stewardship values of the farmers in the Blanchard River watershed. Comparatively, farmers who preferred the All-in-One models were concerned more about the stewardship values that the BMPs represent, while those who preferred single trading models paid more attention to the economic issues such as the cost of money, yield, and land, as well as the subsidies they would receive. The issue of least concern to both groups was “what my neighbors would think about it”. Welch and Marc-

Aurele (2001) and Nowak (2009) found that peer pressure was more influential for late adopters of BMPs; the late adopters were affected by early adopters and neighbors. In the

Blanchard River watershed, the adoption rates of many BMPs were high (e.g. 90% for no-till and conservational tillage, 78% for soil testing, 63% for cover crops and cropping down), suggesting that many farmers were early rather than late adopters. Therefore, the peer pressure was less influential for them. This may be related to the fact that NW Ohio was an early target for the

USDA NRCS program called Lake Erie Conservation Enhancement Program (CREP) which started around 2000 and which required conservation measures be adopted long-term (Richards, 2004).

Figure 2.7 Selection of BMPs for farmers who preferred “All-in-One” Model.

Cover Crops No-till WQT Variables Odds Ratio P Odds Ratio P Odds Ratio P AllinOne1 1.0624 0.9186 0.2388 0.1617 9.4121 0.0432 T.Concern 0.9887 0.7349 1.0427 0.4629 1.2749 0.0103 TotalUsing 0.9271 0.5300 1.3121 0.2334 1.1088 0.5768 TillRank2 2.0167 0.4906 0.7416 0.8841 1.7220 0.6871 TillRank3 1.9498 0.5202 5.6916 0.3468 29.5475 0.0529 TillRank4 1.5546 0.6932 3.1677 0.5939 11.5883 0.1915 CoverCrop1 6.0302 0.0093 1.1451 0.9143 0.7462 0.7805 Age 1.0127 0.5371 1.0625 0.1783 1.0260 0.3253 Gender1 0.9289 0.9354 71843449 0.9919 9.8159 0.3982 Education2 0.5330 0.4080 2.2490 0.5201 1.5603 0.7135 Education3 0.4943 0.3528 0.0749 0.0961 0.4120 0.5577 Education4 0.3208 0.3256 166.5008 0.0524 319.2581 0.0807 Live_yr 0.9916 0.4182 1.0102 0.5502 1.0083 0.6182 Own 1.0047 0.3673 0.9942 0.2725 1.0055 0.4701 Lease 1.0055 0.2869 0.9896 0.0496 1.0063 0.3955 Land.1 0.9956 0.3745 1.0096 0.0680 0.9953 0.5096 Leasep 1.9962 0.6822 60.0994 0.1851 52.4573 0.1928 Manage_yr 1.0102 0.3097 1.0080 0.5818 0.9750 0.0924 Table 2.4 Results of Binary Logistic Regression (presented in odds ratio).

Cover Crops No-till WQT Variables Odds Ratio P Odds Ratio P Odds Ratio P Income2 0.3663 0.2157 0.0095 0.0056 0.0137 0.0037 Income3 0.6494 0.7593 0.0028 0.0872 1099097 0.9940 Income4 0.5054 0.5203 0.0162 0.0656 0.0427 0.0824 Income5 0.3849 0.3755 0.0475 0.0840 0.1027 0.1699 Income_role2 0.8725 0.8686 6.6525 0.2145 5.0886 0.1958 Income_role3 1.1613 0.8696 3.3401 0.3849 205.6139 0.0108 Income_role4 0.5004 0.4682 0.7901 0.8723 1.3415 0.8372 Knowledge1 2.3594 0.2076 9.7571 0.0535 2.0440 0.5352 Table 2.4 Continued.

Figure 2.8 Comparison of farmers with different trading model preference regarding issues about BMP adoption.

2.3.5 Farmers’ Interest in WQT

The third hypothesis tested in this study was that the All-in-One trading model would

increase farmers’ interest in WQT participation. The Binary Logistic Regression results showed

that there was a significant positive relationship (p=0.0432) between the preference for All-in-

One model and farmers’ interest in WQT (Table 2.4). As other variables were held constant, the odds of those who preferred the All-in-One model to be interested in WQT participation was

9.41 times higher than those who preferred models other than All-in-One. Although the

scientific and institutional soundness of WQT projects per se does not necessarily lead to increased willingness to participate (Breetz, 2005), a sophisticated project design that corresponds to farmers’ values is still critical. The All-in-One model intrigues farmers’ interest in

WQT, this may due to the potential of higher economic return, as well as the holistic view of ecosystem provided by the All-in-One model. A further understanding of the determinants of farmers’ interest in WQT would be helpful in the design of an All-in-One program.

This study found that farmers’ interest in WQT was significantly related to their concerns about local water quality issues, household income and the role of farming in the total household income. The positive impact of environmental consciousness on farmers’ BMP adoption was found in multiple studies, e.g. Haghjou et al. (2014), Reimer et al. (2014),

Gedikoglu and McCann (2012). In this study, as the water quality concern score, an index of environmental consciousness, increased by 1 unit, the odds of a farmer becoming interested in

WQT increased by 1.27 times, when other variables were fixed. As farmers become more aware of the environmental problems, they are more motivated by non-economic incentives (Perry-

Hill and Prokopy 2014; Ryan et al., 2003 and Greiner et al., 2009). In this case, if WQT was

introduced as merely a financial incentive, it might not be able attract the attention of farmers

with higher environmental consciousness, a group with a greater possibility to participate.

Farmers valued their roles as environmental stewards; having steward intentions could

positively influence BMP adoption (Tiwari et al., 2008; Chouinard et al., 2016). Therefore, environmental benefits and stewardship values should also be emphasized in WQT programs; the All-in-one model can be designed with this in mind.

The total household annual income and the contribution of farming to total household income had significant impacts on farmers’ interest in WQT participation. The low income group was significantly less interested in WQT participation, while households whose farming income contributed 51% to 90% of the total income were significantly more interested than other groups (Table 2.4). Similar results were found by Welch and Marc-Aurele (2001) that financial tools could better motivate farmers who received income mostly from farming; Xie

(2014) suggested that farmers with higher farming incomes were usually associated with larger scale farms of corn and soybean in Ohio, thus the economic variable played a more important role. In addition, larger farm operators were more willing and had more resources to invest in new technology (Haghjou et al., 2014; Pannell et al., 2014; Prokopy et al., 2014). It is notable that, in this study, households which received over 90% of household income from farming did not support WQT as much as the 51-90% group. The latter group had at least two income sources: farming and non-farming incomes; farming was the main source of income, while at least one family member had a non-farming job. These households usually had enough motivation to sustain the farm, at the same time, they also had a non-farming income as a

backup or as a buffer to risks that might potentially affect their farming income if they

participated in WQT. This finding suggested that in addition to economic and environmental benefits, the risks of increasing cost and/or losing other benefits are of concern by farmers; any

WQT projects should take into account risk perception and management in its design.

Farmers also rated their degree of concern over issues related to WQT. In general, farmers

who were interested in WQT showed a higher concern score than those not interested in WQT

(Figure 2.9), suggesting the former group wanted to learn more about WQT and expected those

issues to be well addressed and explained. The issue of greatest concern for both groups was

the potential of increasing regulation if they participated in WQT. This might be the major

barrier preventing participation, as the farmers who were not interested in WQT scored higher

in this question than the interested group, while all other questions scored lower (Figure 2.9).

Who determines the amount of credits was also a top-rated question. In fact, the question

“how much I trust who recommended it (a certain BMP)” was also an issue of high concern.

According to Breetz (2005) and Mariola (2009), having a trusted organizer/broker is a key to

increase farmers’ participation in environmental programs such as WQT. The most trusted

agency in Blanchard River watershed was the Hancock County Soil and Water Conservation

District, followed by Ohio Farm Bureau, and OSU Extension (Figure 2.10). Other agencies,

including NRCS, the EPA, and the Great Lakes Commission were much less trusted. A survey to

find out the most trusted organizer/broker among stakeholders is necessary for any WQT

program.

For farmers supporting WQT, the third issue which concerned farmers most was the possible improvement that WQT programs could make, meanwhile, issues regarding payment, such as who receives payment, how and when the payment would be made, even the price of credit and trading ratio, were less of concern. This finding echoes the earlier finding that financial incentives alone were not enough to encourage participation, and is supported by

Moore (2000) and Breetz (2005). Again, it is important to emphasize both economic and environmental benefits in BMP adoption programs; this is especially true for the Blanchard

River watershed.

Figure 2.9 Comparison of farmers with different trading model preference in concerned issues about WQT.

Figure 2.10 Farmers’ Trust in Agencies for WQT.

2.4 Discussion

Although most WQT research focuses on market mechanisms, understanding the factors influencing farmers’ adoption of BMPs is also of critical importance to increase participation in

WQT programs (Shortle, 2013). However, these studies are quite limited, partly because the adoption of BMPs is a complicated process that involves many variables. As summarized by

Knowler and Bradshaw (2007), Prokopy et al. (2008) and Liu et al. (2018), there are at least 60 explanatory factors that have been studied by different empirical research around the world.

Following Liu et al. (2018), these factors could be categorized into 1) information and awareness of BMPs; 2) external incentives/disincentives, including financial incentives, social norms, peer pressure, and macro factors e.g. location, climate and policy; 3) characteristics of farmers, including demographics, knowledge, attitudes, perception of risk and uncertainties,

time preferences, and environmental consciousness; 4) characteristics of farms, including the geophysical, sociopolitical and management characteristics such as farm size, soil fertility, land tenure, conservation program enrollment, etc.; 5) characteristics of BMPs, such as observability, ease of use, time requirement, cost-effectiveness; and 6) interactions among

BMPs, e.g. the adoption of one BMP affects the adoption of other BMPs. The influences of

these variables were found significant for some BMPs while insignificant for others; the

influence could also be positive, negative or unclear. What makes it more complicated is that

the adoption is a temporally dynamic process; farmers could be motivated by different factors

at different stages of adoption (Pannell and Vanclay, 2011; Rogers, 2010; Pannell, 1999). The

adoption of BMPs could not be explained by a universal set of factors and there is no consistent

influence of the studied factors (Knowler and Bradshaw, 2007; Prokopy et al., 2008). Also, one study cannot exhaust all potentially influencing factors. However, studying the impact of certain factors on the adoption of certain BMPs is meaningful to guide conservation work on a local

scale. Conservation programs designed according to local conditions are more likely to be

supported by local community.

To increase WQT participation, some studies suggested limiting the trading program to a

small, bilateral scale (Woodward, Kaiser, and Wicks, 2002). Moore (2014) testified in a

Congressional hearing titled "The Role of Water Quality Trading in Achieving Clean Water

Objectives" that, according to the experience of the Alpine Cheese Nutrient Trading Plan, small-

scale community-based WQT programs provide more benefits over larger-scaled programs;

WQT programs should target minor rather than large-scale NPDES permit holders, because the

price of wastewater treatment facility upgrades for small town facilities can be as much as

seven times higher than the cost for larger cities. On a small, local scale, WQT allows the community members, both the credit sellers and buyers, to work together to solve problems while both benefit economically and ecologically from the trading program. At the local level opportunities will exist for culturally adding value to ecosystem services through religious, aesthetic, recreational, or inspirational values that humans derive from ecosystems (Moore,

2014).

WQT is mostly regarded as an external economic incentive to promote BMPs. Mixed results of the economic incentives were found by different studies (Rode et al., 2014). Financial payments might undermine the intrinsic motivations for preserving public goods such as water and soil (Andrews, 2013); cost sharing programs could both undermine and reinforce the adoption of cover crops, conservation tillage and contour-strip (Fleming et al., 2015). As found in this study, economic benefits alone were not able to significantly motivate cover crops and no-till adoption, or encourage WQT participation, because environmental benefits were also valued by farmers in Blanchard River watershed. However, the All-in-One model was found favored by farmers and could significantly increase farmers’ interest in WQT. Although the All- in-One model or the “credit stacking” was originally designed to make WQT more appealing by providing more economic returns, it also delivers a concept that different ecosystem services can be provided by one more holistic ecosystem coupling natural and human biocomplexity.

Compared to conventional WQT that focuses on one singular credit, e.g. phosphorus, nitrogen, or carbon, credit stacking has the potential to better reflect the integrity of ecosystems. Given that environmental consciousness was significant to WQT participation and environmental

benefits were highly valued by the Blanchard farming community, the All-in-One model that

60 provides environmental benefits in addition to economic benefits could be effective in motivating WQT participation and BMP adoption.

2.5 Conclusion

Farmers indicated a clear preference for the All-in-One credit stacking WQT model. This model could significantly increase farmers’ interest in WQT participation, but this does not imply that they also will adopt cover crops or no-till plowing. The All-in-One model provides a higher potential of economic returns, as well as an integrated view of the ecosystem. In the

Blanchard River watershed, farmers’ interest in WQT was related to both environmental concerns and income supporting the substantivist position that people are motivated by a combination of economic and non-economic factors; they valued both environmental and economic benefits when considering participating in WQT and adopting BMPs. Therefore, a type of credit stacking WQT model I have termed “All-in-One is promising in meeting the

Blanchard farmers’ moral, cultural, and economic needs. As noted by Knowler and Bradshaw

(2007), Prokopy et al. (2008) and Liu et al. (2018), the reasons for conservation practice adoption are varied. The adoption of cover crops and no-till is quite complex, because BMP adoption is a dynamic process that involved many variables, including noneconomic ones and consists of different stages. Although the All-in-One trading model was associated with farmers’ intentions to adopt additional cover crops or no-till plowing, this study has identified several determinants for the adoption of the two BMPs. Farmers tended to continue and expand the use of cover crops if they had already experimented with it on their farms. Farmers with lower household income and/or those who operated on larger scale using leased farmland were less

likely to adopt no-till plowing. The difference in the adoption rates might also be explained by the different adoption stages such as having prior experimenting experience of the two BMPs.

Given the complexity of the BMP adoption process and favoring both economic and non- economic rationale for adoption a localized, community-based All-in-One trading project may

have more promise of success in the Blanchard River watershed. This would also be consistent with the successful experience of small-scale community-based WQT projects such as Alpine

Cheese Nutrient Trading Plan (Moore, 2014) or South Nation project in Ontario Canada.

References

Allred, S. B., & Ross-Davis, A. (2011). The drop-off and pick-up method: An approach to reduce

nonresponse bias in natural resource surveys. Small-Scale Forestry, 10(3), 305-318.

Andrews, A.C., Clawson, R.A., Gramig, B.M., Raymond, L. (2013). Why do farmers adopt

conservation tillage? An experimental investigation of framing effects. J Soil Water Conserv

68, 501-511.

Brandt, B. & Baird J. (2008). BMP Challenge: Yield and Income Risk Protection for Corn Farmers

Who Adopt Water Quality BMPs. Presentation. Retrieved from:

http://www.dep.state.pa.us/dep/subject/advcoun/ag/2008/August2008/BMP%20Challeng

e%20PA%20.pdf

Breetz, H. L., Fisher-Vanden, K., Jacobs, H., & Schary, C. (2005). Trust and communication:

Mechanisms for increasing farmers’ participation in water quality trading. Land Economics,

81(2), 170-190.

Carroll, N., & Jenkins, M. (2013). The matrix: Mapping ecosystem service markets. Ecosystem

Marketplace. Retrieved from: http://www.ecosystemmarketplace.com/wp-

content/uploads/2015/09/the_matrix.pdf

Chouinard, H.H., Wandschneider, P.R., Paterson, T. (2016). Inferences from sparse data: An

integrated, meta-630 utility approach to conservation research. Ecol Econ 122, 71-78.

Cooley, D., & Olander, L. (2011). Stacking ecosystem services payments: Risks and solutions.

Nicholas Institute for Environmental Policy Solutions, Working Paper NI WP, , 11-04.

Costanza, R., d’Arge, R., De Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K., Naeem, S.,

O’Neill, R. V., & Paruelo, J. (1998). The value of ecosystem services: Putting the issues in

perspective. Ecological Economics, 25(1), 67-72.

Costanza, R., de Groot, R., Sutton, P., van der Ploeg, S., Anderson, S. J., Kubiszewski, I., Farber, S.,

& Turner, R. K. (2014). Changes in the global value of ecosystem services. Global

Environmental Change, 26, 152-158.

Daily, G. (1997). Nature's services: Societal dependence on natural ecosystems Island Press.

Eugene, P. O. (1962). Relationships between structure and function in the ecosystem. Japanese

Journal of Ecology, 12(3), 108-118.

Fleming, P., Lichtenberg, E., Newburn, D.A. (2015). Agricultural Cost Sharing and Conservation

Practices for Nutrient Reduction in the Chesapeake Bay Watershed, 2015 AAEA & WAEA

Joint Annual Meeting, July 26-28, San Francisco, California. Agricultural and Applied

Economics Association & Western Agricultural Economics Association.

Fox, J. (2008). Getting two for one: Opportunities and challenges in credit stacking.

Conservation and Biodiversity Banking: A Guide to Setting Up and Running Biodiversity

Credit Trading Systems.

Gedikoglu, H., McCann, L.M. (2012). Adoption of win-win, environment-oriented, and profit-

oriented practices among livestock farmers. J Soil Water Conserv 67, 218-227.

Green, D., & Shapiro, I. (1996). Pathologies of rational choice theory: A critique of applications

in political science. Yale University Press.

Greiner, R., Patterson, L., Miller, O. (2009). Motivations, risk perceptions and adoption of

conservation practices by farmers. Agr Syst 99, 86-104.

Haghjou, M., Hayati, B., Choleki, D.M. (2014). Identification of Factors Affecting Adoption of Soil

Conservation 664 Practices by Some Rainfed Farmers in Iran. J Agric Sci Technol 16, 957-

967.

Hess, M. (2004). ‘Spatial’relationships? towards a reconceptualization of embeddedness.

Progress in Human Geography, 28(2), 165-186.

King, D. M., & Kuch, P. J. (2003). Will nutrient credit trading ever work? an assessment of supply

and demand problems and institutional obstacles. Environmental Law Reporter News and

Analysis, 33(5), 10352-10368.

Knowler, D. & Bradshaw, B. (2007). Farmers’ adoption of conservation agriculture: A review and

synthesis of recent research. Food Policy, 32, 25-48.

Kramer, J. (2003). Lessons from the trading pilots: Applications for Wisconsin water quality

trading policy. Resource Strategies, Inc., Madison, WI,

Lindner, J. R., Murphy, T. H., & Briers, G. E. (2001). Handling nonresponse in social science

research. Journal of Agricultural Education, 42(4), 43-53.

Liu, T., Bruins, R. J., & Heberling, M. T. (2018). Factors Influencing Farmers’ Adoption of Best

Management Practices: A Review and Synthesis. Sustainability, 10(2), 432.

Mariola, M. J. (2009). Are Markets the Solution to Water Pollution? A Sociological Investigation

of Water Quality Trading,

McSweeny, W. T., & Kramer, R. A. (1986). The integration of farm programs for achieving soil

conservation and nonpoint pollution control objectives. Land Economics, 62(2), 159-173.

Melevin, P. T., Dillman, D. A., Baxter, R. K., & Lamiman, C. E. (1999). Personal delivery of mail

questionnaires for household surveys: A test of four retrieval methods. Journal of Applied

Sociology, 69-88.

Methorst, R., Roep, D., Verstegen, J., & Wiskerke, J. S. (2017). Three-fold embedding: Farm

development in relation to its socio-material context. Sustainability, 9(10), 1677.

Millennium Ecosystem Assessment. (2005). Ecosystems and human wellbeing: A framework for

assessment. Millennium Ecosystem Assessment. Washington, DC: Island Press.

Moore, R. H. (2009). Ecological Integration of the Social and Natural Sciences in the Sugar Creek

Method. Sustainable agroecosystem management. Taylor and Francis Group, CRC, 21-40.

Moore, R. H., (2014). The role of trading in achieving water quality objectives: Congress

testimony before the Water Resources and Environment Subcommittee Committee on

Transportation and Infrastructure United States House of Representatives. Washington D.C.

Movafaghi, O.S., K. Stephenson, and D. Taylor. (2013). Farmer Response to Nutrient Credit

Trading Opportunities in the Coastal Plain of Virginia. Paper presented at the annual

meeting of the Agricultural and Applied Economics Association, Washington, D.C., U.S.,

August 4-6.

National Research Council (NRC). (2001). Compensating for wetland losses under the Clean

Water Act. Washington D.C.: National Academy Press.

North Carolina Ecosystem Enhancement Program. (2009). Nutrient offset program. Raleigh N.C.

Nowak, P. (2009). The subversive conservationist. J Soil Water Conserv 64, 113A-115A.

Pannell, D.J. (1999). Social and economic challenges in the development of complex farming

systems. Agroforest Syst 45, 395-411.

Pannell, D.J., Llewellyn, R.S. & Corbeels, M. (2014). The farm-level economics of conservation

agriculture for 766 resource-poor farmers. Agric Ecosyst Environ 187, 52-64.

Pannell, D.J. & Vanclay, F. (2011). Changing land management: Adoption of new practices by

rural landholders. CSIRO Publishing

Perry-Hill, R. & Prokopy, L. (2014). Comparing different types of rural landowners: Implications

for conservation practice adoption. J Soil Water Conserv 69, 266-278.

Polanyi, K., & MacIver, R. M. (1944). The great transformation(Vol. 2, p. 145). Boston: Beacon

Press.

Popkin, S. L. (1979). The rational peasant: The political economy of rural society in Vietnam.

Univ of California Press.

Prokopy, L.S., Floress, K., Klotthor-Weinkauf, D., Baumgart-Getz, A. (2008). Determinants of

agricultural best management practice adoption: Evidence from the literature. J Soil Water

Conserv 63, 300-311.

Prokopy, L.S., Towery, D., Babin, N. (2014). Adoption of Agricultural Conservation Practices:

Insights from Research and Practice, Purdue Extension.

Reimer, A., Thompson, A., Prokopy, L.S., Arbuckle, J.G., Genskow, K., Jackson-Smith, D., Lynne,

G., McCann, L., Morton, L.W., Nowak, P. (2014). People, place, behavior, and context: A

research agenda for expanding our understanding of what motivates farmers' conservation

behaviors. J Soil Water Conserv 69, 57A-61A.

Richards, R.P. (2004). Ohio Lake Erie CREP Program: Annual Report on Water Quality. Retrieved

from: http://wwwapp.epa.ohio.gov/dsw/nps/NPSMP/docs/2004HLECREP_Report.pdf

Robertson, M., BenDor, T. K., Lave, R., Riggsbee, A., Ruhl, J., & Doyle, M. (2014). Stacking

ecosystem services. Frontiers in Ecology and the Environment, 12(3), 186-193.

Rode, J., Gómez-Baggethun, E., Krause, T. (2014). Motivation crowding by economic incentives

in conservation policy: A review of the empirical evidence. Ecol Econ 109, 80-92.

Rogers, E.M., (2010). Diffusion of innovations. Simon and Schuster.

Ryan, R.L., Erickson, D.L., De Young, R. (2003). Farmers' motivations for adopting conservation

practices along 801 riparian zones in a mid-western agricultural watershed. J Environ

Planning Manage 46, 19-37.

Scott, J. C. (1977). The moral economy of the peasant: Rebellion and subsistence in Southeast

Asia. Yale University Press.

Shaffer, S., & Thompson, E., Jr. (2013). Encouraging California specialty crop growers to adopt

environmentally beneficial management practices for efficient irrigation and nutrient

management: Lessons from a producer survey and focus groups. American Farmland Trust,

p. 26.

Shortle, J. (2013). Economics and Environmental Markets: Lessons from Water-Quality Trading.

Agricultural and Resource Economics Review, 42, 5774.

Steele, J., Bourke, L., Luloff, A., Liao, P., Theodori, G. L., & Krannich, R. S. (2001). The drop-

off/pick-up method for household survey research. Community Development, 32(2), 238-

250.

Tiwari, K.R., Sitaula, B.K., Nyborg, I.L.P., & Paudel, G.S. (2008). Determinants of farmers'

adoption of improved soil conservation technology in a Middle Mountain watershed of

Central Nepal. Environ Manage 42, 210-222.

U.S. Department of Agriculture (USDA). (2012). Census of agriculture. National Agricultural

Statistics Service. Retrieved from:

https://www.agcensus.usda.gov/Publications/2012/index.php.

U.S. Environmental Protection Agency (USEPA). (2004). Water quality trading assessment hand-

book: EPA region 0’s guide to analyzing your watershed. Washington D.C.

Van Maasakkers, M. (2016). The creation of markets for ecosystem services in the United States:

The challenge of trading places Anthem Press.

Welch, E.W. & Marc-Aurele Jr, F.J. (2001). Determinants of farmer behavior: adoption of and

compliance with best management practices for nonpoint source pollution in the

Skaneateles Lake watershed. Lake Reservoir Manage 17, 233-245.

Willamette Partnership. (2010). Ecosystem credit accounting. Willamette Partnership. Retrieved

from: http://willamettepartnership.org/ecosystem-credit-accounting.

Woodward, R. T., Kaiser, R. A., & Wicks, A. B. (2002). The structure and practice of water quality

trading markets. JAWRA Journal of the American Water Resources Association, 38(4), 967-

979.

Xie, Y. (2014). Watershed modeling, farm tenancy and adoption of conservation measures to

facilitate water quality trading in the upper scioto watershed, ohio The Ohio State

University.

Zhong, H., Qing, P., & Hu, W., (2015). Farmers' willingness to participate in best management

practices in Kentucky. J Environ Planning Manage, 1-25.

Chapter 3. Water Quality Trading for On-site Septic System Nutrient Management under the

Changing Climate Conditions

3.1 Introduction

In general, water quality trading (WQT) is an approach that allows point sources, such as

factories and wastewater treatment plants, to meet their regulatory obligations (e.g. National

Pollutant Discharge Elimination System (NPDES) Permit) by using pollutant reductions created

by another source, such as agriculture, who has lower pollution control costs (USEPA, 2004).

Although water quality trading projects usually have the same ultimate goal of reducing

pollution on a watershed scale, in practice, the design and focus of each project varies from

each other. For example, the multi-state water quality trading project in the Ohio River Basin

led by EPRI emphasized creating a nutrient credit trading market on a larger scale (Willamette

Partnership, 2012); on the other hand, the South Nation River project in Ontario, Canada,

tended to be a community-based local project where the price of credits was only one of

several considerations (O’Grady, 2011).

In the South Nation River WQT program, which inspired many projects after its 2006

presentation at the 2nd National Water Quality Trading Conference, septic system upgrades

was their top conservation project. However, upgrading septic systems have noticeably been

absent from other water quality trading projects, probably owing to the high cost per pound of

phosphorus generated and the emphasis on agricultural conservation measures. US EPA (2009) had identified the challenges of trading using septic systems: 1) septic systems’ contribution to

water quality impairment is low compared to other sources such as agriculture; 2) for credit buyers, the costs of pursuing trading with septic systems may compromise the benefits; 3) the

distance of septic systems to a waterbody may vary; 4) wastewater treatment plants provide a more cost effective target with a higher nutrient reduction; 5) the areas where septic tanks are located may be developed in the future where they would be replaced by wastewater treatment plants. However, some programs, such as the Chesapeake Bay trading (Maryland

Water Quality Trading Advisory Committee, 2017) and the Montana Nutrient Trading program

(Walsh, 2014), considered incorporating septic systems into their trading schemes because of the environmental and public health impacts of septic systems.

Given its small scale and dispersed distribution, on-site septic systems are often assumed to have limited contribution in nutrients, especially when compared to agriculture (Withers et al.,

2014). Also, due to insufficient data on their patterns of nutrient emission, septic systems have long been ignored in watershed management programs. However, studies have found that the effluent of septic systems could have a large impact on local water quality and stream ecology

(Steffy and Kilham, 2004; Palmer-Felgate et al., 2010). For example, a study using high-

resolution monitoring of Ireland and Northern Ireland rural watersheds has shown that clusters

of malfunctioning septic systems can be a major source of nutrients, especially in spring and summer periods (Macintosh et al. 2011).

In Ohio, the Ohio Department of Health estimated about 352 tons/yr of TP (total phosphorus) was contained in the on-site septic system effluent of the 148,000 homes discharging to the Lake Erie watershed (ODH, 2008). About 25% of this discharge reaches a

waterway, contributing an estimated 88 tons/yr of TP to Lake Eire. In the Blanchard River

Watershed, a subwatershed of the Maumee River watershed that drains into the Western Lake

Erie Basin, septic systems are the 4th largest source of phosphorus, contributing 7.83 tons of

phosphorus to the watershed every year, higher than the contribution of point sources (6.19

ton/yr), (Ohio EPA, 2009).

In addition, public health concerns have also brought septic systems into the spotlight.

Septic systems are the major reservoir of human enteropathogens, poorly treated septic system effluent can contaminate groundwater and surface water, threatening public health. For example, Borchardt et al (2003) found that children’s endemic diarrheal illness in central

Wisconsin was associated with septic system density. Children were infected by drinking well water that was contaminated by septic system effluent. In the case of the Blanchard River

Watershed, where a former large marsh was located, high water table and frequent flooding coupled with failed septic systems has the potential to lead to massive pathogen dispersion.

On-site septic systems became a pollution source for multiple reasons. In Ohio, the main causes of septic system malfunctioning include aging, overloading, soil limitations and site limitations (Table 3.1). Soil limitations are the fundamental issue, because septic systems’ ability to remove sewage pollutants largely relies on the natural process of soil. Soil is the best medium to treat wastewater; an adequate layer of soil can remove suspended solids, organic matters, ammonia, bacteria and viruses (Mancl, 2013). The three most important characteristics of soil for effective wastewater treatment are, 1) an unsaturated condition to

73 promote aerobic condition; 2) a soil layer deeper than 4 feet (Figure 3.1); and 3) soil permeable to air and water (Mancl, 2013). In addition, soil pores must be fine enough to trap pathogens.

Principle Reasons for Failure

System Aging 44%

Direct discharge exceeds limits 43%

Soil Limitations 33%

Site Limitations 25%

System Owner Abuse 17%

Design Issues 14%

No Leach Field 14%

Unapproved System 7%

Installation Issues 3%

Other 1%

Table 3.1 Principal Reasons for Septic System Failure (adopted from ODH & Ohio EPA, 2013).

Soil surface

Infiltration Infiltration Suspended solids removal: 1 foot

Ammonia, Organic matter, bacteria removal: 1.5 to 2 feet

Virus removal 0.5 to 2 feet following organic matter removal

Figure 3.1 Pollutant removal soil depths for wastewater infiltrating unsaturated soil (adopted from Mancl and

Slater, 2013).

These characteristics are limited by seasonal saturation, dense till, fragipan, bedrock, flooding, slope, permeability and other conditions of different soil types (Mancl, 2013).

According to the soil types in Ohio, Mancl and Slater (2001) have classified the soil into three categories: 1) soil series suited for traditional leach lines systems or mound systems, 2) soil series suited for mound systems only; and 3) soil series not suited for soil-based wastewater treatment. The classification is based on soil depth to bedrock, depth to restrictive layer, depth to seasonal high water table, soil permeability and soil permeability at soil surface, which were measured in the National Cooperative Soil Survey (1960-2000). Soil types that are considered not suitable for soil-based wastewater treatment are those that are hydric, less than 4 feet in depth above water table or a restrictive layer, or subject to frequent flooding, or poorly permeable soil (Mancl and Slater, 2001). Sixty-eight percent soil in Ohio is considered unsuitable for treating wastewater and the occurrence of soil unsuitable for soil-based septic systems is the highest in northwest Ohio (Mancl and Slater, 2001), where the former Great

Black Swamp located. The Great Black Swamp was one of the biggest wetlands in the U.S. a century ago. Although it was transformed into farmland with deep artificial ditches, a large portion of soil in this area remains easily saturated. For example, about 40% of soil in Hancock

County, Ohio, a county located within the former Great Black Swamp, is hydric soil (USDA Web

Soil Survey, 2005). Figure 3.2 shows the characteristics of the most commonly found soil type in

Hancock County. In addition to being rated as hydric soil, it is also poorly drained, shallow to water table and frequently ponding, disqualifying this soil type from being suitable for septic system leach field.

Characteristics of PmA • Slope: 0 to 1 percent • Depth to restrictive feature: More than 80 inches • Natural drainage class: Very poorly drained • Depth to water table: About 0 to 12 inches • Frequency of flooding: None • Frequency of ponding: Frequent • Hydric soil rating: Yes Figure 3.2 PmA (Pewamo silty clay loam) is the most common soil type in Hancock County, accounting for 24.4% of the total area of Hancock County.

Besides Mancl and Slater, the USDA Web Soil Survey (2005) rated the suitability of soil

for septic systems based on the cumulative effect of multiple limiting conditions. In Hancock

County, Ohio, 98% of the soil is ranked “very limited” for the use of septic tank absorption fields, indicating that “the soil has one or more features that are unfavorable for the specified use.

The limitations generally cannot be overcome without major soil reclamation, special design, or expensive installation procedures. Poor performance and high maintenance can be expected.”

Considering the importance of soil limitations in effective wastewater treatment, the state law of Ohio is clear regarding the need for qualified soil specialists to review each septic tank

siting. However, it appears that in practice the local soil types are not well-aligned with septic system proper functioning. For example, previous work by Xiaoping Wei (2012) found that human fecal coliforms in poorly treated septic system effluent became the primary source of microbial contamination in Upper Sugar Creek watershed in northeast Ohio, mainly due to the use of septic systems with traditional soil adsorption leach fields whose effectiveness was limited by the local soil conditions.

Given the environmental and public health impacts, it is critical to address the failing septic systems, especially in northwest Ohio where the soil conditions are limiting the effectiveness of pollutant removal. This chapter investigates the possibility of promoting septic system upgrades using water quality trading as a part of the community-based watershed management in the

Blanchard Watershed in northwest Ohio. This watershed is a part of the Maumee River watershed, a major source of the nutrients that cause eutrophication and Harmful Algae Bloom

(HAB) in Western Lake Erie Basin. Following water quality trading programs that use a community-based approach such as the South Nation (O’Grady, 2011), Alpine Cheese and

Muskingum plans (Moore, 2014), incorporating rural household septic systems upgrade into water quality trading has the potential of broadening community engagement within a county.

County health departments or soil and county water conservation districts could create septic plans that could be placed in bidding systems such as that of the Great Miami Conservancy

Water Quality Trading Plan in Ohio. Having septic systems included in the water quality trading programs also has the added benefit of bridging all rural residents--both farmers and nonfarmers-- to address the environmental and public health issues as a community. In this chapter, we aimed to 1) understand rural residents’ interest in water quality trading for septic system

77 upgrades, 2) identify households and their locations for a pilot project. Our hypothesis was that water quality trading has the potential to serve as an incentive for septic system upgrades.

3.2 Methodology

3.2.1 Study Area

The Blanchard River Watershed in northwest Ohio is a part of the Maumee River Watershed that drains into the Western Lake Erie Basin (Figure 3.3). It is a HUC8 watershed that covers

493,415 acres spanning across 5 counties: Hancock, Hardin, Putnam, Allen and Wyandotte in

Ohio. This area is a part of the former Great Black Swamp, one of the biggest wetlands in the

U.S. a century ago. The Great Black Swamp was drained artificially by deep ditches and transformed into an agricultural landscape. Due to the legacy of the swamp and the low elevation, cities like Findlay and Ottawa in downstream Blanchard River are frequently flooded.

Moreover, the Maumee River watershed scored as one of the highest watersheds in the Corn

Belt for having experienced flooding over the past five years and highest for farmer concern about the increased loss of nutrients into waterways and increasingly frequent extreme rains

(Loy et al 2013).

This study focused on the overlapping area of Blanchard River Watershed and Hancock

County, Ohio. About 3/4 of this county falls within Blanchard River Watershed, taking up over

50% of area in the watershed. According to US Census Bureau (2017), Hancock County had a population of 75,573 and 31,083 households in 2015; 94.3% of the population are white, 1.7%

are Asian, 1.5% are African American, 3.4% are other races; 91.6% of the population older than

25 are high school graduates or higher. The total area of Hancock County is 340,070 acres, including 230,261 acres of farmlands (2012 Census of Agriculture). Of the farmland, 92.4% is

cropland; in 2012 the total crop sales were $146,881,000 and livestock sales were $13,364,000.

The County seat is the City of Findlay located in the northern part of the County.

(a) (b)

Figure 3.3 (a) Blanchard River Watershed (adopted from USDA, 2011); (b) Study area.

3.2.2 Data Collection

Survey

A questionnaire survey was used to understand the current conditions of septic systems

and rural residents’ interests in upgrading their septic system for water quality trading. The

sample frame was all rural households within the boundary of the Hancock County part of the

Blanchard River Watershed. Eligible households were those that 1) had an on-site septic system;

2) were within the boundary of the Blanchard River Watershed; and 3) were within the

boundary of Hancock County. A total of 1,891 households met these criteria. These households

were divided into two groups: 541 farming households and 1,300 non-farming households.

Farming households completed the septic system survey and a farming survey, while non-

farming households completed only the septic system survey. The addresses of these

households were obtained from the Hancock County Auditor for survey delivery. The “Drop-

Off/Pick-Up” method (Melevin et al. 1999) was used for this survey. This method has the

advantage of reducing the nonresponse bias (Steele et al. 2009), especially in natural resource

surveys (Allred et al. 2010). The face-to-face communication also allows the researcher to

better determine the eligibility of the respondent and gain experiential insights (Steele et al.

2009, Allred et al. 2010).

Each member of the research team (including the primary author and three local student assistants) who administered the survey was certified by the Collaborative Institutional Training

Initiative (CITI) for human subjects and the survey was approved by the Ohio State University

Institutional Review Board of the Office of Responsible Research. From October 2016 to

February 2017, we visited 900 households in the study area, including all the 541 farming households and 359 randomly selected non-farming households. Among the 900 households,

578 had no one home in the first and second visit, 138 had refused to participate in the survey; among the 184 who completed the surveys, 96 were farming households and 88 were non-

farming households. The final response rate was 57.1% (Table 3.2).

Farming Household Non-Farming Household Total Completed 96 88 184 Refused 53 85 138 Response Rate 64.4% 50.9% 57.1% Table 3.2 Responses of septic system survey, from both farming and non-farming households.

Public records and GIS dataset

The data used for this study was also from public records. The two main sources of public data were the Hancock County Auditor database and online GIS datasets. The County Auditor provided the information of residential houses, including age, area, value, tax and number of rooms of the houses, etc. The GIS dataset include soil type, soil depth, elevation, and slope from USDA, USGS and local agencies. Using GIS, we also measured the distance between residences and the nearest stream.

3.2.3 Analytical Approach

Binary logistic regression models were employed to investigate the relation between willingness to upgrade the septic system under three different scenarios: 1) an increased intensity rainfall scenario, 2) an updated regulation scenario, and 3) a WQT scenario.

Independent variables included status of septic systems, environmental perceptions, concerns over regulation, demographic, house and geographic characteristics (Table 3.3). Kernel density analysis using ArcGIS was employed to study the spatial distribution of households with different willingness to upgrade septic systems and key characteristics.

Variables Label Descripti Coding on Interest in septic system upgrades • Interest in upgrades under YRain intensified rainfall scenario 0: Not interested • Interest in upgrades under 2015 YReg Binary 1: Interested (any degree of interest in any regulation scenario of the three trading models) • Interest in upgrades under WQT YWQT scenario Status of septic systems • System Age NewSystem Binary 0: System installed before 1990s (out of lifespan) 1: System installed after 1990s (within lifespan)

• Maintenance in the past 5 years Maintenance 4-class 0: No maintenance ordinal 1: Adopted 1 maintenance practice (inspect/pump out/use treatment products) 2: Adopted 2 maintenance practices 3: Adopted 3 maintenance practices

• Perceived effectiveness of septic TotalRemove Numeric Combination of the ratings of effectiveness system in removing sewage, in sewage, pathogen and nutrients removal pathogen and nutrients Range: 0-15 (0= not effective, 25= most effective)

Environmental perceptions • Perceived water quality in Quality 6-class 0: No idea nearby streams ordinal 1: Poor 2: Fair 3: Average 4: Good 5: Excellent

• Environmental concerns over TotalConcern Numeric Combination of the concern scores of the local water quality issues issues of bacteria in streams, pathogens in flood water, well water contamination, fish population decline, and toxic algae bloom in Lake Erie. Range: 0-25 (0= not concerned, 25= most concerned)

• Awareness of the contribution TotalContribu Numeric Combination of the perceived contribution of septic systems to local water tion of septic systems to the bacteria in quality issues streams, pathogens in flood water, well water contamination, fish population decline, and toxic algae bloom in Lake Erie. Range: 0-25 (0= no contribution, 25= most contribution) Table 3.3 Variables included in the survey willingness to upgrade septic systems.

Variables Label Description Coding Concerns over increased regulation on RegulationCo 5-class 0: No idea septic system ncern ordinal 1: Not concerned 2: Slightly concerned 3: Somewhat concerned 4: Very concerned 5: Excellent

Demographic characteristics Farmer Farmer Binary 0: Not Farmer 1: Farmer Income Income 3-class 1: Low income (<50k/yr) ordinal 2: Medium income (50k-100k/yr) 3: High income (>100k/yr) People live in current house People Numeric Years live in current house YearsLive Numeric

House characteristics Age of house HouseAge Numeric No. of bedrooms Room Numeric Living area LivingArea Numeric Total Property value TotalValue Numeric

Geographical characteristics Soil Natural Drainage Class SoilDrainage 4-class 1: Somewhat poorly drained ordinal 2: Very poorly drained 3: Moderately well drained 4: Well drained Elevation Elevation Numeric Distance to stream Stream 4-class 1: within 50m from a stream ordinal 2: within 100m from a stream 3: within 150m from a stream 4: farther than 150m from a stream Table 3.3 Continued.

3.3 Results and Discussion

3.3.1 Current Status of Septic Systems in the Blanchard River Watershed

The 2013 Ohio Household Sewage Treatment Systems and Failures Report (ODH & Ohio EPA,

2014) estimated the overall residential septic system failure rate in Ohio was approximately

31%. The estimated failure rate in northwest Ohio, where the Blanchard River Watershed located, was the highest (39%) comparing with other parts of the state. The report also

identified principal reasons of septic system failure (Table 3.1). Aging is the most common cause of failure. The life span of a septic system is usually between 20 to 30 years. Our survey found that 55% of the sampled septic systems were installed after 1990, which were within the life span (Figure 3.4). The rest were mostly out of date. The age of septic systems is related to the age of houses. Older houses tend to have older systems. It is notable that in this area 45% of the sampled houses were over 100 years old. These old houses were scattered throughout the study site (Figure 3.5).

Figure 3.4 Septic systems installation year.

Figure 3.5 Distribution of households using septic systems (the darker color shows older systems).

Soil limitation is also causing septic system malfunctioning. Pewamo is the most common soil type in Hancock County. The soil types where the septic systems located were determined by overlapping the locations of the responded households with the USDA National Cooperative

Soil Survey map. Silt loam is the most commonly found soil in the 0 to 1 foot depth, while silty clay is most common in the 1 to 4 feet depth (Table 3.4). According to the USDA Soil Survey,

65.87% of the soil at the locations of responded households was classified as "somewhat poorly drained" and 13.49% was classified as "very poorly drained", suggesting nearly 80% of the septic systems were located on soil that was not suitable for removing nutrients and pathogens.

Soil type Natural Drainage Class

0 to 1 Feet Deep 1 to 4 Feet Deep

Clay 9.15% Somewhat poorly drained 61.27%

Clay loam 1.41% 14.08% Very poorly drained 17.61%

Fine sandy loam 2.11% 0.70% Moderately well drained 18.31%

Loam 19.72% 4.93% Well drained 2.82%

Silty clay 69.01%

Silty clay Loam 15.49% 1.41%

Silt loam 61.27% 0.70%

Table 3.4 Soil types and natural drainage class at the locations of the responded household s.

Although the septic system were aged and limited by soil conditions, responding

households were active in maintaining their septic systems. In the past 5 years, 40.72% of the

septic systems were inspected, 65.87% were pumped out, and 38.92% were treated with

treatment products. Only 17.40% of the residents did not do anything to maintain their septic

systems (Figure 3.6). This suggests that many residents in this area were responsible septic system users. Moreover, the average household size of the responded households was 2.43 persons, while the average number of bedrooms was 3.32. Since the capacity of septic systems is designed according to number of bedrooms, in most cases, the amount of sewage produced by the responded households should be within the designed capacity. However, an overload could happen during certain periods of time, such as holidays when the houses accommodate more people than usual or in severe weather events that cause flooding and soil saturation.

Unfortunately, routine maintenance alone is not able to overcome the problems caused by system deteriorating tanks, under-sized systems and soil limitations. Septic system upgrades

are still necessary.

100%

90%

80%

70% 65.87%

60%

50%

40.72% 40% 38.92%

30%

20% 15.57%

10%

4.19% 2.99% 2.99% 0% Pumping Inspection Treatment No Action Repair Pipe Replacement Tank Replacement

Figure 3.6 Maintenance of septic systems in the past 5 years.

3.3.2 Perceived Environmental Quality and Risk by Households in the Blanchard River

Watershed

A majority of the responding households were confident in the effectiveness of their septic

systems: most respondents rated their septic systems to be "quite effective" and "very

effective" in removing sewage, bacteria and nutrients; less than 2% rated them as "not

effective" (Figure 3.7, 3.8, 3.9). Generally, respondents had more confidence in sewage removal effectiveness than in bacteria and nutrients removal effectiveness. Residents using a newer system tended to be more confident than those using an older system. Households were asked to rate the water quality in the stream nearest to their houses based on their own judgment:

15.3% rated their stream water quality as "excellent," 36% as "good," 20% as "average," 7.1% as "fair," 5.3% as "poor," while 16.5% had "no idea." Thus, overall, 51% rated the water quality of their local stream as either “excellent” or “good”.

Figure 3.7 Perceived sewage removal effectiveness.

Figure 3.8 Perceived bacteria/virus/pathogen removal effectiveness.

Figure 3.9 Perceived nutrients removal effectiveness.

3.3.3 Households’ Willingness to Upgrade Septic Systems

In this study, households’ willingness to upgrade their septic systems was assessed in three

scenarios: 1) increased frequency and intensity of heavy rain events in the area, causing more

severe water pollution; 2) a new state regulation regarding septic systems in January 2015; 3)

the concept of water quality trading was presented with three trading models. The three

models were: a) the return from participating in water quality trading was not specified; b) the

participating households would receive an annual payment of $50/household as return; c)

instead of payment, the trading fund would be used to hire a local professional to manage the

participating households’ septic systems.

In the intensified rainfall scenario, households were provided information about the impact

of climate change on local water quality:

Studies have shown that at least 70% of nutrients (e.g. phosphorus and nitrogen) runoff

happens during a few heavy rain storms (heavy rains are those more than 0.3 inch/hour).

Evidence show that the number of heavy rains in Ohio has increased in the past decade, and

this trend is likely to continue by 2050. This suggests that more nutrients would be loaded to

streams under future climate conditions. Will you consider upgrading your septic system to

prevent nutrient runoff?

In the 2015 regulation scenario, households were asked whether they had considered upgrading their septic systems after the new rules regarding septic systems were adopted in

Ohio since January 2015.

Although the majority had no intention to upgrade their septic systems in either scenario, the percentage of those having the intention to upgrade regarding intensified rainfall (33.55%)

was higher than those having the intention regarding the new regulation (12.50%) (Figure 3.10 a-b). This could be explained by the finding that households were very concerned (55.69%)

about more regulations being put upon their septic systems (Figure 3.11).

(a)

(b)

Figure 3.10 Households’ willingness to upgrade septic systems in (a) the scenario of intensified rainfall and (b)

the scenario of 2015 regulation.

Figure 3.11 Households’ concern about regulation regarding septic systems.

Figure 3.12 showed that the responded households were much more willing to upgrade

their septic systems in the scenario of WQT than in the other two scenarios (intensified rainfall or 2015 regulation). The households showed no preference to any of the three WQT models

(Table 3.5): 43.21% had some degree of interest in the general idea of water quality trading,

42.14% were interested in the annual payment model, and 43.48% were interested in the

professional management model. Overall, 58.07% households showed some degree of interest

in at least one of the three trading models for septic system upgrades. O’Grady (2011) of the

South Nation River WQT project, which had successfully incorporated septic system upgrades

into its scheme, pointed out that community agreement was a critical condition for a WQT

project to succeed. In the Blanchard River watershed, the idea of WQT was accepted by a

majority of rural households, suggesting its potential to serve as an incentive for septic system

92 upgrades. It should also be noted that, since WQT is still a new concept in the area, more effort should be made to communicate the idea to the local community on a broader scope.

Figure 3.12 Households’ willingness to upgrade septic systems under three scenarios.

Not Slightly Somewhat Very Trading Scenarios Interested Interested Interested Interested Interest in the general idea of WQT 56.79% 27.78% 14.20% 1.23%

Interest in WQT with an annual 57.86% 30.19% 10.69% 1.26%

payment

Interest in WQT with a professional 56.52% 29.19% 12.42% 1.86%

management plan

Overall interest in any of the above 41.94% 41.29% 14.84% 1.94%

three WQT scenarios

Table 3.5 Households’ interests in septic system upgrades under the scenarios of three WQT models.

The spatial analysis showed the clustering of households that were likely to upgrade septic

systems under the intensified rainfall scenario, 2015 regulation scenario and WQT scenario

(Figure 3.13a-c). The overlapping area of households willing to upgrade in each scenario, was located around the upstream area of Blanchard River main stem and Lye Creek tributary within

Jackson and Delaware Township of Hancock County, Ohio, suggesting that this area might have greater potential for a pilot rural wastewater management project than other areas in the watershed.

(a)

(b)

Figure 3.13 Geographical distribution of households’ willingness to upgrade septic systems in (a) the intensified

rainfall scenario, (b) the 2015 regulation scenario and (c) the WQT scenario.

(c)

Figure 3.13 Continued.

Figure 3.14 showed the geographical distribution of households interested in septic system upgrades under the three WQT models. Households that were most interested in the general idea of WQT were concentrated near the upstream area of Blanchard River and

Lye Creek, mostly located in Jackson and Delaware Township (Figure 3.14a). The same

pattern was found for the interest in annual payment trading model (Figure 3.14b) and

interest in professional management trading model (Figure 3.14c). The distribution of old

septic systems (Figure 3.5) did not exactly match that of households with high interest in

WQT, but it is notable that there are old systems concentrated near the upstream area of

Blanchard River and Lye Creek, indicating that this area also in urgent need of septic system upgrades.

(a)

(b)

Figure 3.14 Geographical distributions of households’ general interest in WQT (a), WQT with an annual payment (b) and WQT with professional management plan (c).

(c)

Figure 3.14 Continued.

3.3.4 Factors Associated with Households’ Willingness to Upgrade Septic Systems

Among all independent variables, perceived effectiveness of septic systems, perceived

water quality in nearby streams, environmental concerns, concerns about governmental

regulation, household income and age of house were significantly related to the households’

willingness to upgrade septic systems (Table 3.6). The perceived septic systems effectiveness in removing sewage, pathogens and nutrients had a negative relationship with willingness to upgrade the system. For the perceived effectiveness to increase by 1 unit, the odds of a household becoming willing to upgrade decreases by 14.10% in the intensified rainfall scenario; the odds deceases by 10.73% in the water quality trading scenario. Households that considered

98 their septic systems to be more effective saw less necessity for upgrading, implying that the limitation of soil condition in this area, as well as the fact that septic systems have a finite lifespan were largely neglected. As discussed above, education of the factors limiting septic system effectiveness is critical in enhancing households’ awareness of the need for system upgrades. The concerns about governmental regulation had a great negative impact on households’ willingness to upgrade in the intensified rainfall scenario. The odds of those who were “somewhat concerned” or “very concerned” about increasing regulation to upgrade is about 95% lower than those had no idea about this issue.

Intensified rainfall scenario Regulation scenario WQT scenario Variables Odds Ratio P Odds Ratio P Odds Ratio P Status of septic systems NewSystem1 1.32657 0.63864 0.15552 0.07330 0.81799 0.67850 Maintenance1 0.24268 0.06109 0.32336 0.38820 0.38880 0.12020 Maintenance2 0.66140 0.58757 0.77051 0.82540 0.65455 0.50160 Maintenance3 0.87695 0.86687 7.37429 0.10530 0.48408 0.30910 TotalRemove 0.85899 0.00691 0.96656 0.70500 0.89270 0.02670 Environmental perceptions Quality1 2.02405 0.52866 71.59319 0.04510 3.03133 0.28890 Quality2 58.38155 0.00330 3.25763 0.58270 14.64355 0.02010 Quality3 4.94314 0.07830 72.74790 0.02130 4.70676 0.04400 Quality4 3.85357 0.11666 11.62316 0.14790 3.30026 0.07810 Quality5 0.25640 0.37321 4.23761 0.47090 1.69317 0.53780 TotalConcern 1.34972 0.00189 1.29486 0.08940 1.10528 0.17160 TotalContribution 0.89333 0.19478 0.78600 0.05030 1.04028 0.54330 Concerns over regulation RegulationConcern1 0.07669 0.13764 258395005 0.99520 0.32400 0.40230 RegulationConcern2 0.10861 0.16994 41450064 0.99560 0.15335 0.15430 RegulationConcern3 0.04532 0.03691 986821766 0.99480 0.19437 0.16630 RegulationConcern4 0.05464 0.04961 470826396 0.99500 0.27665 0.25090 Table 3.6 Results of binary logistic regression for households’ willingness to upgrade septic systems in three scenarios (presented in odds ratio).

Intensified rainfall scenario Regulation scenario WQT scenario Variables Odds Ratio P Odds Ratio P Odds Ratio P Demographic characteristics Farmer1 0.46320 0.17614 0.29229 0.14600 0.68530 0.41910 Income2 0.33354 0.11188 0.50652 0.47130 0.59055 0.29800 Income3 3.91964 0.05000 0.28280 0.25940 3.58945 0.03940 People 0.90186 0.69354 1.07929 0.86780 1.48364 0.08870 YearsLive 0.97449 0.22203 1.03539 0.28010 1.02147 0.23100 House characteristics HouseAge 1.02057 0.00230 1.01000 0.28550 0.99846 0.76970 Room 0.81269 0.55832 0.69226 0.46260 0.56479 0.05170 LivingArea 1.00827 0.34810 1.00390 0.75270 1.00595 0.37030 TotalValue 1.00000 0.70681 1.00000 0.41430 1.00000 0.83100 Geographical characteristics SoilDrainage2 0.46269 0.23567 1.51437 0.70400 0.65968 0.49470 SoilDrainage3 0.84434 0.83422 3.21234 0.42560 0.48773 0.30390 SoilDrainage4 0.55399 0.71719 11.76348 0.27050 1.14557 0.91600 Elevation 0.99913 0.67871 1.00291 0.33730 0.99962 0.82760 Stream2 0.09283 0.25003 0.00000 0.99320 0.20495 0.34020 Stream3 0.11567 0.30635 0.00000 0.99150 0.40438 0.58780 Stream4 0.09710 0.20299 0.02105 0.07430 0.45145 0.58720 Table 3.6 Continued.

The perceived water quality in nearby streams was positively related to upgrade willingness in all three scenarios. For households that considered the local water quality to be “poor”, the odds of being willing to upgrade was 71.59 times higher than those had no idea about water quality in the regulation scenario. For those who considered water quality to be “fair”, the odds of being willing to upgrade was 58.38 times higher (versus those had no idea about water quality) in the intensified rainfall scenario and 14.64 times higher in the water quality trading scenario. For those who considered water quality to be “average”, the odds of being willing to upgrade was 72.75 times higher in the regulation scenario and 4.71 higher in the water quality trading scenario. Moreover, the odds of households that considered water quality to be “good”

100

or “excellent” to be willing to upgrade were lower. Households’ concern over environmental

issues was also an important factor. As 1 unit increased in the environmental concern score, the

odds of a household being willing to upgrade the septic system increased by 1.35 times, in the

intensified rainfall scenario. The more a household was concerned about the local aquatic

environment, the more likely they would upgrade the septic system. As found in other studies

(e.g. Prokopy, Floress, Baumgart-Getz, & Klotthor-Weinkauf, 2008; Moore et al., 2016; Morton

et al., 2016; Prokup, Wilson, Zubko, Heeren, & Roe, 2014), the awareness of local environment

degradation and concern of environment quality had a positive effect on behavior change.

Future rural household wastewater management programs should focus on the education of

local environmental issues and fostering environmental awareness.

High income households were also more likely to upgrade septic systems. In the intensified

rainfall scenario, the odds of high income (annual income > $100k) households being willing to

upgrade was 3.92 times higher than the low income households (annual income < $50k); in the

water quality trading scenario, the odds was 3.59 times higher. Cost is a major prohibiting

factor in septic system upgrades. According to the Ohio Department of Health (2008), in

northwest Ohio, the average cost of installing a septic system with shallow leach lines was

$ 7,988 and $11,355 for a sand mound system. The high cost of replacing or upgrading a septic

system could present a significant financial hardship to the low income households. In the study

area, for instance, 19.4% of the households had less than $50,000 annual income. Financial

support programs are necessary to help these households. Currently, Community Block

Development Grants/Community Housing Improvement Program (CBDG/CHI), U.S. Department

of Agriculture, Rural Development, 502 Home Ownership, 504 Home Repair Grants and Loans,

101

Federal Housing Administration HUD funds and the Ohio Housing Trust Fund and some other

sources are available. However, many of these grants are either highly competitive or restricted

to households with certain eligibilities (Ohio Department of Health, 2008). More funding

opportunities should be made available to households that need septic system upgrades or

replacement.

3.4 Conclusion

The failing rate of household septic systems in northwest Ohio where the Blanchard watershed located was 39%, the highest in Ohio (Ohio Department of Health & Ohio

Environmental Protection Agency, 2013). The inadequately treated household wastewater from malfunctioning septic systems becomes a source of nutrients that cause Lake Erie HAB and threaten public health. Failure is largely caused by old and poorly sited systems. Soil in the

former Great Black Swamp in northwest Ohio is wet and poorly drained, resulting in poor

performance of regular soil-based septic systems. Since routine maintenance is unable to

overcome these challenges, septic system upgrades are needed. However, most watershed

management programs have failed to address the failing septic system issue. The feasibility of

incorporating septic system upgrades in a WQT program was considered. Most (58.07%) of the

responding households in the Blanchard River watershed were willing to upgrade their septic

systems in a WQT program, which is much higher than the upgrade willingness in the

intensified rainfall (33.55%) or new regulation scenarios (12.50%). WQT has the potential to

serve as an incentive for septic system upgrades. The households that were willing to upgrade

are clustered in the upstream area of Blanchard River main stem and Lye Creek tributary within

102

Jackson and Delaware Township of Hancock County, Ohio. Pilot projects are likely to work well in this area.

For a septic system upgrade program to succeed in northwest Ohio, it should focus on the following aspects:

1) Resident education regarding the impacts of soil limitations on septic systems performance. Septic systems like other home fixtures that have a finite life expectancy. Old septic systems were designed by a different standard than the modern systems and system components deteriorate over time. The limitations of soil could not be overcome with regular system and routine maintenance. Those households that better understood the limitation of system effectiveness were more likely to upgrade their septic systems.

2) Enhanced awareness of local environment degradation and concern for environment quality has a positive effect on households’ willingness to upgrade septic systems. Households that perceived local water quality to be “fair” or “average” and those more concerned about local environmental issues are more willing to upgrade. Education should have a local focus and relate households to their nearby aquatic environment.

3) Financial support should be made available for septic system upgrades. Given that septic system replacement or upgrades are expensive, the high income households (annual income >

$ 100k) have significantly higher willingness to upgrade their septic systems comparing with the low income households (annual income < $ 50k). Some funding sources for septic system are available, but programs should make sure these opportunities are known and accessible for households that need them.

103

Based on the finding that certain groups of households were more willing to upgrade their septic system in water quality trading, future studies may focus on the design and implementation of these programs.

References

Allred, S. B., & Ross-Davis, A. (2011). The drop-off and pick-up method: An approach to reduce

nonresponse bias in natural resource surveys. Small-Scale Forestry, 10(3), 305-318.

Borchardt, M. A., Chyou, P. H., DeVries, E. O., & Belongia, E. A. (2003). Septic system density

and infectious diarrhea in a defined population of children. Environmental Health

Perspectives, 111(5), 742-748.

Green, J. E. (2002). Evaluating phosphorus migration from septic systems near otsego lake. 34th

Ann.Rept.(2001).SUNY Oneonta Biol.Fld.Sta., SUNY Oneonta.

Ho, J.C., & Michalak, A.M. (2017). Phytoplankton blooms in Lake Erie impacted by both long-

term and springtime phosphorus loading. Journal of Great Lakes Research, 43(3), 221-228.

Kane, D.D., Conroy, J.D., Richards, R.P., Baker, D.B., & Culver, D.A. (2014). Re-eutrophication of

Lake Erie: Correlations between tributary nutrient loads and phytoplankton biomass.

Journal of Great Lakes Research, 40(3), 496-501.

Levy, S. (2017). Learning to love the great black swamp. UNDark. Retrieved from:

https://undark.org/article/great-black-swamp-ohio-toledo/

104

Loy, A., Hobbs, J., Arbuckle Jr, J., Morton, L., Prokopy, L., Haigh, T., Knoot, T., Knutson, C., Mase,

A., & McGuire, J. (2013). Farmer perspectives on agriculture and weather variability in the

corn belt: A statistical atlas. Cscap 0153-2013,

Macintosh, K., Jordan, P., Cassidy, R., Arnscheidt, J., & Ward, C. (2011). Low flow water quality

in rivers; septic tank systems and high-resolution phosphorus signals. Science of the Total

Environment, 412, 58-65.

Mancl, K., & Slater. B. (2013). Suitable of Ohio soils for treating wastewater. OSU Extension

Bulletin.

Maryland Water Quality Trading Advisory Committee. (2017). Maryland trading and offset

policy and guidance manual Chesapeake Bay watershed. Maryland Department of

Environment. Retrieved

from:http://mde.maryland.gov/programs/water/Documents/WQTAC/TradingManualUpda

te4.17.17.pdf

McDonald, J. H. (2009). Handbook of biological statistics Sparky House Publishing Baltimore,

MD.

Meehan, H. (2004). Phosphorus migration from a near-lake septic system in the otsego lake

watershed, summer 2003. 36th Annual Report (2003).SUNY Oneonta Bio.Fld.Sta., SUNY

Oneonta,

105

Melevin, P. T., Dillman, D. A., Baxter, R. K., & Lamiman, C. E. (1999). Personal delivery of mail

questionnaires for household surveys: A test of four retrieval methods. Journal of Applied

Sociology, , 69-88.

Michalak, A.M., Anderson, E.J., Beletsky, D., Boland, S., Bosch, N.S., Bridgeman, T.B., Chaffin,

J.D., Cho, K., Confesor, R., Daloglu, I., Depinto, J.V., Evans, M.A., Fahnenstiel, G.L., He, L., Ho,

J.C., Jenkins, L., Johengen, T.H., Kuo, K.C., Laporte, E., Liu, X., McWilliams, M.R., Moore,

M.R., Posselt, D.J., Richards, R.P., Scavia, D., Steiner, A.L., Verhamme, E., Wright, D.M., &

Zagorski, M.A. (2013). Record-setting algal bloom in Lake Erie caused by agricultural and

meteorological trends consistent with expected future conditions. Proceedings of the

National Academy of Sciences of the United States of America, 110(16), 6448-6452.

Moore, R.H., (2014). The role of trading in achieving water quality objectives: Congress

testimony before the Water Resources and Environment Subcommittee Committee on

Transportation and Infrastructure United States House of Representatives. Retrieved from

https://transportation.house.gov/uploadedfiles/2014-03-25-moore.pdf

Moore, R. H., Lekies, K., Todey, D., Miller, W., Blockstein, D., Higgins, T., Nkongolo, N.,

Abendroth, L.J., & Morton, L. W. (2016). Agri-climate education: Preparing the next

generation. Technical report series: Observations and recommendations of the climate and

corn-based cropping systems coordinated agricultural project. Iowa State University, Ames,

IA.

106

Morton, L.W., Prokopy, L.S., Arbuckle, J.G., Ingels, C. Jr., Thelen, M., Bellm, R., Bowman, D.,

Edwards, L., Ellis, C., Higgins, R., Higgins, T., Hudgins, D., Hoorman, R., Neufelder, J.,

Overstreet, B., Peltier, A., Schmitz, H., Voit, J., Wegehaupt, C., Wohnoutka, S., Wolkowski,

R., Abendroth, L., Angel, J., Haigh, T., Hart, C., Klink, J., Knutson, C., Power, R., Todey, D.,

and Widhalm. M. (2016). Climate change and agricultural extension: Building Capacity for

land grant extension services to address the agricultural impacts of climate change and the

adaptive management needs of agricultural stakeholders. In Technical report series:

Findings and recommendations of the climate and corn-based cropping systems

coordinated agricultural project. Iowa State University, Ames, IA.

National Oceanic and Atmospheric Administration. (2016). U.S. Climate Extremes Index.

Retrieved from www.ncdc.noaa.gov/extremes/cei.

Natural Resources Conservation Service. (2008). Rapid watershed assessment-data profile

Blanchard River watershed. Retrieved from: http://wleb.org/leadership/LdrshpMtgs/2008-

04-03%20Handouts.pdf

O’Grady, D. (2011). Sociopolitical conditions for successful water quality trading in the south

nation river watershed, Ontario, Canada. JAWRA Journal of the American Water Resources

Association, 47(1), 39-51.

Ohio Department of Agriculture, Ohio Department of Natural Resources, Ohio Environmental

Protection Agency & Ohio Lake Erie Commission. (2013). Ohio Lake Erie phosphorus task

107

force II final report. Retrieved from:

http://lakeerie.ohio.gov/Portals/0/Reports/Task_Force_Report_October_2013.pdf

Ohio Department of Health. (2008). ODH report to the Ohio Lake Erie phosphorous task force.

Retrieved from:

http://www.epa.ohio.gov/portals/35/lakeerie/ptaskforce/041808ODHoleptfRPT.pdf

Ohio Department of Health. (2013). Household sewage treatment system failures in Ohio: A

report on local health department survey responses for the 2012 clean watersheds needs

survey. Ohio Department of Health. Retrieved from: https://www.odh.ohio.gov/-

/media/ODH/ASSETS/Files/eh/STS/12HSTSSystemsandFailures.pdf?la=en

Ohio Environment Protection Agency (Ohio EPA). (2009). Total maximum daily loads for the

Blanchard river watershed. Retrieved from:

http://www.epa.state.oh.us/portals/35/tmdl/blanchardrivertmdl_final_may09_wo_app.pd

Ohio Environment Protection Agency (Ohio EPA). (2010). Ohio Lake Erie phosphorus task force

final report. Retrieved from:

http://www.epa.state.il.us/water/nutrient/presentations/lake_erie_task_force.pdf

Ohio Environment Protection Agency (Ohio EPA). Ohio nutrient reduction strategy. Retrieved

from: http://epa.ohio.gov/dsw/wqs/NutrientReduction.aspx#146064466-nutrient-strategy

108

Palmer-Felgate, E. J., Mortimer, R. J., Krom, M. D., & Jarvie, H. P. (2010). Impact of point-source

pollution on phosphorus and nitrogen cycling in stream-bed sediments. Environmental

Science & Technology, 44(3), 908-914.

Prokopy, L.S., Floress, K., Baumgart-Getz, A., & Klotthor-Weinkauf, D. (2008). Determinants of

agricultural best management practice adoption: Evidence from the literature. Journal of

Soil and Water Conservation, 63(5), 300-311.

Prokup, A., Wilson, R., Zubko, C., Heeren, A, and Roe, B. (2017). 4R nutrient stewardship in the

Western Lake Erie Basin. Columbus, OH: The Ohio State University, School of Environment

& Natural Resources.

Steele, J., Bourke, L., Luloff, A., Liao, P., Theodori, G. L., & Krannich, R. S. (2001). The drop-

off/pick-up method for household survey research. Community Development, 32(2), 238-

250.

Steffy, L. Y., & Kilham, S. S. (2004). Elevated δ15N in stream biota in areas with septic tank

systems in an urban watershed. Ecological Applications, 14(3), 637-641.

Stumpf, R.P., Wynne, T.T., Baker, D.B., & Fahnenstiel, G.L. (2012). Interannual variability of

cyanobacterial blooms in Lake Erie. PLOS One, 7(8), e42444.

The Freshwater Trust. (2016). Uplift report. Retrieved from

https://www.thefreshwatertrust.org/wp-content/uploads/2017/10/TFT-Uplift-Report-

2016-web.pdf

109

U.S. Census Bureau. (2017). Quick Facts: Hancock County, Ohio. Retrieved from:

https://www.census.gov/quickfacts/fact/table/hancockcountyohio/AGE275210

U.S. Department of Agriculture (USDA). (2005). Web Soil Survey. Retrieved from:

https://websoilsurvey.nrcs.usda.gov/

U.S. Department of Agriculture (USDA). (2011). Western Lake Erie basin watershed map.

Retrieved from

https://www.nrcs.usda.gov/wps/portal/nrcs/detail/oh/programs/?cid=nrcs144p2_029639

U.S. Department of Agriculture (USDA). (2012). Census of agriculture. Retrieved from:

https://www.agcensus.usda.gov/Publications/2012/index.php.

U.S. Environmental Protection Agency (USEPA). (1980). Design manual: Onsite wastewater

treatment and disposal systems. Washington, DC.

U.S. Environmental Protection Agency (USEPA). (2002). Onsite wastewater treatment systems

manual. Washington, DC.

U.S. Environmental Protection Agency (USEPA). (2004). Water quality trading assessment hand-

book: EPA region 0’s guide to analyzing your watershed. Washington D.C.

U.S. Environmental Protection Agency (USEPA). (2007). Water quality trading toolkit for permit

writers Appendix F: Trading with subsurface septic systems. Retrieved from:

https://www3.epa.gov/npdes/pubs/wqtradingtoolkit_app_f_trading_septic.pdf

110

Van Liere, K. D., & Dunlap, R. E. (1981). Environmental concern: Does it make a difference how

it's measured? Environment and Behavior, 13(6), 651-676.

Vedachalam, S., Hacker, E., & Mancl, K. (2012). The evolution of septic systems practices in Ohio.

Journal of Environmental Health, 75(5), 22-7.

Vollmer-Sanders, C., Allman, A., Busdeker, D., Moody, L. B., & Stanley, W. G. (2016). Building

partnerships to scale up conservation: 4R nutrient stewardship certification program in the

lake erie watershed. Journal of Great Lakes Research, 42(6), 1395-1402.

Walsh, J., Wuebbles, D., Hayhoe, K., Kossin, J., Kunkel, K., Stephens, G., Thorne, P., Vose, R.,

Wehner, M., Willis, J., Anderson, D., Doney, S., Feely, R., Hennon, P., Kharin, V., Knutson, T.,

Landerer, F., Lenton, T., Kennedy, J., & Somerville, R. (2014). Our changing climate: Climate

change impacts in the United States. In The Third National Climate Assessment.

Washington, D.C.

Walsh, M. J. (2014). Water quality trading business case for Montana. Montana Department of

Environmental Quality. Retrieved from:

https://deq.mt.gov/Portals/112/Water/WQPB/Standards/NutrientWorkGroup/PDFs/Wate

rQualityTradingBusinessCaseMontana.pdf

Wei, X. (2012). Identification and remediation of microbial contaminants in the headwaters of

an agricultural watershed The Ohio State University.

111

Willamette Partnership. (2012). In it together: A how-to reference for building point-nonpoint

water quality trading programs designing and operating a trading program (part 2 of 3).

Willamette Partnership. Retrieved from: http://willamettepartnership.org/wp-

content/uploads/2014/09/In-It-Together-Part-2_2012-07-30.pdf

Withers, P. J., Jordan, P., May, L., Jarvie, H. P., & Deal, N. E. (2014). Do septic tank systems pose

a hidden threat to water quality? Frontiers in Ecology and the Environment, 12(2), 123-130.

Xie, Y. (2014). Watershed modeling, farm tenancy and adoption of conservation measures to

facilitate water quality trading in the Upper Scioto watershed, Ohio The Ohio State

University.

112

Chapter 4. Climate Change and BMP Adoption in the Blanchard River Watershed, Ohio

4.1 Introduction

While temperature has been included in water quality trading in cases where water is used to cool water from power generating plants, to date the effect of climate change warming on nutrient flows has not entered the discussion on water quality trading. In large part, water quality trading programs have been based on the Total Maximum Daily Loads (TMDL) measured by USEPA. These TMDLs usually are usually calculated during one summer testing a number of sites in a watershed and tend to focus on low flow rather than high flow events. According to

The Third National Climate Assessment (Walsh, 2014) by the U.S. Global Change Research

Program, the agriculture of the U.S. will be affected by more variable and extreme weather in the future, including more frequent heat waves, increased average precipitation, increased heavy rain events, flooding and droughts, and a longer frost-free crop season. In the northern part of the U.S., where the Corn Belt is located, the rainfall amount in very heavy rain events has increased 30-39% from 1958 to 2012 and will continue to increase by up to 5 times by 2081 to 2100 (compared with 1981-2000) (Walsh, 2014). This change will result in substantial great impacts on the soil and water resources as well as the agricultural productivity (Walsh, 2014).

Taking adaptive actions to moderate the effects of climate change is critical to reducing the vulnerability of agriculture. However, studies have detected multiple challenges preventing farmers from taking adaptive actions.

113

The first challenge has been “climate change skepticism” and/or “climate change denial”. A

survey of 5000 farmers from 22 watersheds in 11 Corn Belt states (Loy et al., 2013) found that,

only 8% of the farmers believed that climate change was occurring and was mostly caused by

human activities, while 31% responded that “there is insufficient evidence to know with

certainty whether climate change is occurring or not” and 4% believed that climate change was

not occurring at all. A 2010 survey in Indiana found that 35% of the farmers thought that

climate change was an “issue invented just to scare people” (Gramig et al. 2013). These beliefs are related to the political and social meanings associated with climate change. The skepticism and/or denial of anthropogenic climate change often leads to inaction, since the skeptics and/or deniers believe humans, including themselves, have little control over this “natural cycle”

(Arbuckle et al. 2013); or there is too much uncertainty about climate change to justify changing their farming practices (Loy et al. 2013, Morton et al. 2017).

The second challenge is farmers’ perception of risk related to climate change. Perceptions of risk are “subjective assessments people use to understand and cope with danger and uncertainties in life” (Loy et al. 2013). Howden et al. (2007) emphasized if farmers do not perceive climate change as a threat, it is not likely they will undertake adaptive actions.

Individuals who are willing to take action generally have higher levels of concern and perceive a greater amount of risk of negative impacts on profit. They also perceive the consequences to be more serious, more local, and more personal (Wilson et al. 2013). The survey by Wilson et al

(2013) in the Maumee watershed of Ohio found that less than 16% of the farmers saw the impact of climate change as extremely serious. Less than 40% of the farmers in 11 Corn Belt states were concerned or very concerned about the hazards of flooding, nutrient runoff and soil

114

erosion induced by climate change, while about half were concerned or very concerned about

weed pressure, insect pressure, crop disease, extreme rains and heat stress. In other words,

over half of the farmers perceived climate change as low or no risk (Loy et al., 2013). In practice,

farmers have to assess and manage risk in their farming business on a daily basis. These risks lie

in production, weather, market fluctuation, regulation changes, and social norm expectations.

Compared to these risks, climate change on a longer-term scale might be less personal and local.

Additionally, research also suggests that the higher the degree of control over a hazard, the

lower perception of risk. However, those who regard climate change as a pure natural cycle

over which humans have little control, do not necessarily have a higher perception of risk nor

the intention to take adaptive action (Loy et al. 2013).

The third challenge is farmers’ lack of trust in environmental or agricultural interest groups.

People’s trust in the capacity of the expert organization and their agencies to fulfill their

mandate has a significant influence on people’s perception of risks, especially a large-scale risk

like climate change (Freudenburg, 1993; Kahan et al., 2011). Moreover, the success of

environmental programs often relies on a local, trusted, embedded intermediary that serves as

the link between these programs and farmers (Mariola 2009). Specifically, prior research has found that people’s level of trust in government agencies, especially EPA and NOAA, and scientists who work for the government was an indicator of support for climate policy; while trust in industry and scientists who work for industry was negatively associated with the support of climate policy (Dietz et al. 2007; Zahran et al. 2006). The results of the Corn Belt survey (Loy et al, 2013) showed that seed and chemical dealers had the most influence on farmers’ decision-making, followed by the Natural Resources Conservation Service (NRCS),

115

while conservation NGOs had the least influence. This finding indicates farmers generally have

more trust in the industry or private sector than in the government or environmental sector

and might be another reason why there is insufficient support for climate policy.

Among all the adaptive actions, no-till and cover crops are of special interest in the Western

Lake Erie Basin watershed, because these two practices have the potential to improve water

quality and prevent erosion by mitigating nutrient leaching and reducing wind and water

erosion (ISUEO 2014). This is particularly important in the Western Lake Erie Basin watershed

where more frequent and intensive precipitation due to climate change is expected. Given the

eutrophication and Harmful Algal Bloom (HAB) problem in Lake Erie, the adaptation of

agriculture to future climate conditions is not only crucial to the agricultural production, but

also to the health of the Lake Erie ecosystem and the millions of people who depend on Lake

Erie as a drinking water source. On a national level, about 35% of cropland in the US was in no-

till farming and less than 4% in cover crops (2012 Agriculture Census). In Ohio, where most of

the Western Lake Erie Basin watershed is located, about 39.8% of cropland was in no-till and

3.3% in cover crops (2012 Agriculture Census). For the 20,712 Ohio farms that adopted no-till,

the average no-till cropland area was 207 acres; for the 6,565 farms that adopted cover crops, the average cover crop area was 54 acres (2012 Agriculture Census). Although multiple programs such as the Environmental Quality Incentives Program (EQIP) were designed to promote adoption of conservation measures including no-till and cover crops, the adoption rate remains low.

This chapter aims to understand, the influence of scientific predictions of local climate change on farmer intentions to adopt conservation measures, especially no-till and cover crops

116

that could help them to adapt to climate change. I am also interested in what factors are

related to farmers’ intention in this scenario. The factor I am particularly interested in is

farmers’ experiences with climate. Morton et al (2017) found that some farmers who had the

perception that there was too much uncertainty about the impact of climate change could not

justify changing current agricultural practices and strategies; the possibility of having this

perception was weakly associated with their recent experience with drought but was not found

to be associated with farmers’ recent experiences with flooding. Since farmers are close observers of climate, a more detailed study that examines not only their experiences with extreme weather events, but also their observation of changes in precipitation and temperature patterns is needed to better explain the relationship between farmers’ personal experience and their willingness to adopt adaptive practices. In this study I hypothesize that 1) farmers who observed changes in climate are more likely to take adaptive actions; 2) farmers who observed similar trends as The Third National Climate Assessment (Walsh et al., 2014) are more likely to adopt additional conservation measures in the future.

4.2 Methodology

4.2.1 Data collection

A survey of Blanchard River Watershed farmers located across the 11 townships in Hancock

County, Ohio, was conducted to study their observations of local climate change and its impact to agriculture and water quality, as well as their willingness to participate in water quality trading programs under the scenario of future climate conditions. The sampling frame was constructed using all the registered addresses located on land 1) with an agriculture code, 2)

117

larger than 50 acres, and 3) within the boundary of the 11 townships of Hancock County that

are in the Blanchard River Watershed. The resulting frame contained 541 addresses. Using the

“Drop-Off/Pick-Up” method (Melevin et al. 1999, Riley 2002), I visited all 541 addresses from

October 2016 to February 2017, but 391 of them had no one available when we visited. Among

the 150 who were asked to fill out the survey, 96 completed it and 53 refused (Table 2.1 in

Chapter 2). The final response rate was 64.7%.

4.2.2 Data Analysis

The relationships between farmers’ actions/intentions and climate observations based on

the National Climate Assessment and other factors were examined with binary logistic

regression. The dependent variables were 1) whether farmers had already taken action (Yaction)

and 2) whether farmers had the intention to adopt additional adaptive actions (Yadd). There are

5 categories of independent variables (Table 4.1):

1) The perception of climate change includes the observation of changes in climate and the

prediction of climate change impact on Lake Erie water quality. This observation of climate

change includes the changes in rain intensity, heavy rain frequency, flooding intensity, flooding

frequency, summer temperature and summer precipitation. In each question, I examined whether the observation matched with the scientific record by the National Climate

Assessment, 0=no, 1=yes. A total score that combines the score of each question was also calculated to reflect the overall fitness of the farmers’ observation with the scientific record.

118

2) Perceptions of water quality issues reflected in the farmers’ concern over the current

situation of water quality on and near their farms. There was a series of 13 questions regarding

different aspects of water quality in the survey. Farmers reported their degree of concern to

each question (0=not at all concerned, to 3= very concerned). In order to perform the binary

logistic regression, the scores of each question were combined to result in a total score that

reflected the overall degree of farmer concern water quality issues.

3) Current farming practices that had been already adopted by farmers. There are 3

variables in this category: use of conservation tillage, use of cover crops, and the total number of conservational practices that are being used on the farm.

4) Information on the farm including the total farming acreage, the percentage of leased land in total farming acreage and the length that the family had been managing the farm.

5) Demographic characteristics including the farmers’ age and education level as well as the total annual income of the household.

Variables Label Description Coding Dependent Variables Action to address climate change YACTION Binary 0: No 1: Yes

Willingness to add more BMPs YADD Binary 0: No 1: Yes Independent Variables Perceptions of Climate Change Observation of climate change Observation Numeric For each observation that matches the National Climate Assessment (NCA) record, mark as 1 point; then combine the results of the 6 questions about farmers’ observations of climate change (range: 0- 6). Table 4.1 Variables for the Binary Logistic Regression.

119

Variables Label Description Coding Prediction of the Lake Erie water quality LakeErie Binary 0: No change or improve 1: change due to climate change Worsen Concerns over water quality issues Total Concern Score T.Concern Numeric Combination of concern ratings of all water quality issues listed in the survey (range: 0-39) Current adoption of conservation measures Tillage TillRank 4-class 1: Conventional tillage ordinal only 2: Conventional and conservation tillage 3: Conservation tillage and no-till 4: No-till only Cover crops CoverCrop Binary 0: No 1: Yes Total number of BMPs currently in use TotalUsing Numeric Combination of adopted conservation measures listed in the survey (range: 0-15) Characteristics of the Farmers Age Age Numeric Gender Gender Binary 0: Female 1: Male Education Education 4-class 1: High School ordinal 2: Some college 3: College graduate 4: Graduate degree Years live in Blanchard River watershed Live_yr Numeric Characteristics of the Farms Land owned Own Numeric Land leased Lease Numeric Total farmland (owned + leased) Land Numeric Percentage of leased land in total farmland LeaseP Numeric Years managing the farm Manage_yr Numeric Farming income Income 5-class 1: Less than $24,999 ordinal 2: $25,000 - $49,999 3: $50,000 - $74,999 4: $75,000 - $99,999 5: More than $100,000 Role of farming income to a household Income_role 4-class 1: < 10% ordinal 2: 10-50% 3: 51-90% 4: > 91% income Table 4.1 Continued.

120

4.3. Results

4.3.1 Farmers’ Observation of Climate Change

Farmers were asked to report their observation of changes in precipitation, flooding, and temperature since the year 2000. About 50% of the farmers indicated they had observed more intense and more frequent rainfall since 2000, while about 40% observed no change and less than 10% observed a decreasing trend. The 2014 US National Climate Assessment (U.S. Global

Change Research Program) reported that in Ohio, the amount of precipitation in very heavy events increased 30-39% from 1958 to 2012, while the frequency of heavy rain events could increase up to 5 times by 2081-2100, compared with 1981-2000. The National Climate Data

Center’s Daily Climate Cooperative archives data also showed that in Northwest Ohio, the precipitation has been increasing from 1951 to 2010; the extreme rain events that exceed 1.25 inches have also been increasing since 1951. This trend was observed correctly by 50% of the farmers in this area (Figure 4.1, 4.2).

121

Figure 4.1 Farmers’ observation of rain intensity change since the year 2000.

Figure 4.2 Farmers’ observation of rain frequency change since the year 2000.

122

Given its geographical characteristics and the legacy of the former Great Black Swamp, flooding is a long time concern in the Blanchard River watershed. Figure 4.3 shows the elevation variation in the Blanchard River watershed. The southern part of Hancock County in red is higher than the northern part, which is downstream in the Blanchard River. The blue area suffers more from flooding than the yellow and red area. More farmers had observed increased flooding frequency, but not so many had observed the increase in the intensity of flooding

(Figure 4.4, 4.5). Recorded floods in the county seat of Findlay occurred in multiple times in history and it has become more frequent in the past decade (Figure 4.6).

Figure 4.3 Elevation of Blanchard River Watershed (Adopted from NRCS Rapid Watershed Assessment - Data Profile Blanchard River Watershed, 2008).

123

Figure 4.4 Farmers’ observation of flood intensity change since the year 2000.

Figure 4.5 Farmers’ observation of flood frequency change since the year 2000.

124

Figure 4.6 Crest Height of the Blanchard River at the City of Findlay 1910-2017 (Data from NOAA National

Weather Service).

The National Climate Data Center’s Daily Climate Cooperative archives data showed an

increasing annual average temperature since the 1970s, but most farmers did not observe any

change in summer temperatures or precipitation, while some observed a decrease in summer precipitation (Figure 4.7, 4.8).

125

Figure 4.7 Farmers’ observation of summer temperature change since the year 2000.

Figure 4.8 Farmers’ observation of summer precipitation change since the year 2000.

126

Based on their observation, farmers were asked to predict the impact of changes in climate

(if any) on the water quality problem of Lake Erie. Most farmers predicted the situation would stay the same (57.83%) while 26.51% predicted that the water quality would become worse and 15.66% predicted the situation would improve (Figure 4.9).

Figure 4.9 Farmers’ prediction of Lake Erie water quality change under future climate conditions.

4.3.2 Farmers’ Action to Address Climate Change

Farmers were asked if they had taken any actions on the changes in climate they had observed. This survey found that 44% farmers had taken some actions to deal with the impacts of climate change, 34% had not done anything, and 22% did not respond. A total of 18 different approaches were named by farmers as actions they had adopted to address climate change.

The most common actions included adding or upgrading drainage tiles, changing planting dates

127

according to the weather forecast, adopting no-till and planting cover crops. Other approaches

included adding buffer strips, selecting more adaptable varieties of crops, removing logs from

streams, and building a greenhouse. Most of these approaches aimed to deal with excessive

rainfall. The Corn Belt survey (Loy et al, 2013) found that 67.6% of the farmers in the Western

Lake Erie basin agreed that “farmers should invest more in agricultural drainage systems to

prepare for increased precipitation”, which was the highest among all surveyed Corn Belt

watersheds. Accordingly, this study which was conducted in 2016, four years after the above

Corn Belt survey was conducted in 2012, provides evidence that many farmers had already taken this into action.

We used binary logistic regression to detect the association between farmers’ action and potentially influential factors (Table 4.1). The results (Table 4.2) showed that, the observation of changes in climate and education were the only factors significantly associated with farmers’ action to deal with climate change. While other variables were held constant, for every one unit of change in the total observation score (one observation that matches the NCA record earns one observation score), the odds of falling into the group of farmers who had already taken action (versus falling into the group of farmers who had not taken action) increased by 1.45 units (Table 4.2). With a closer examination, I found that the observation of summer precipitation decrease was the most critical to farmers’ taking action (Table 4.3). Those who had observed decreased summer precipitation were 4.62 times more likely to take action than those who had observed no change in summer precipitation. The observation of increased flood frequency is also an important factor, with a P value (0.0654) close to significant level.

128

YACTION Variables Odds Ratio P

Observation 1.4482 0.0431

LakeErie1 1.9416 0.4131

T.Concern 1.0570 0.1635

TillRank2 3.3636 0.2319

TillRank3 1.2921 0.8034

TillRank4 2.0512 0.5300

CoverCrop1 0.3537 0.1501

TotalUsing 1.2871 0.0721

Age 0.9948 0.8053

Gender1 2.9987 0.3803

Education2 0.7809 0.7475

Education3 0.1281 0.0176

Education4 3.4076 0.4252

Live_yr 0.9947 0.6669

Own 0.9927 0.1239

Lease 0.9944 0.2258

Land 1.0062 0.1717

LeaseP 0.1818 0.3082

Manage_yr 1.0006 0.9582

Income2 1.7221 0.5174

Income3 0.7354 0.8517

Income4 1.7590 0.6290

Income5 1.6558 0.6570

Income_role2 0.2223 0.1171

Income_role3 0.5288 0.5346

Income_role4 0.7273 0.7629

Table 4.2 Binary logistic regression results for farmers’ action and associated variables (presented in odds ratio).

129

YACTION Variables Odds Ratio P

rain.intensity1 1.4642 0.5299

rain.frequency1 0.3501 0.1327

flood.intensity1 1.4576 0.5668

flood.frequency1 3.6590 0.0654

summer.rain1 4.6214 0.0195

summer.temp1 0.6018 0.4681

Table 4.3 Binary logistic regression results for farmers’ action and observation of rain, flood and temperature (presented in odds ratio).

4.3.3 Farmer’s Intention to Adopt Additional BMPs under Future Climate Conditions

Farmers were provided with the information about the impacts of climate change on local water quality and asked the following question about whether they would add more BMPs to their farms:

Studies have shown that at least 70% of nutrients (e.g. phosphorus, nitrogen) runoff

happens during a few heavy rain storms (heavy rains are those more than 0.3 inch/hour).

Evidence shows that the number of heavy rains in Ohio has increased in the past decade,

and this trend is likely to continue by 2050. This suggests that more nutrients would be

loaded into streams under future climate conditions. Will you consider adding MORE

conservation measures to your farm to prevent nutrient runoff?

About 77% of the farmers indicated they would add more BMPs (Figure 4.10). This is substantially higher than the finding of the farmers’ survey by Corn Belt Survey (Loy et al, 2017),

130

where 59.2% of the farmers in the western Lake Erie Basin agreed that “I should take additional

steps to protect the land I farm from increased weather variability.”

Figure 4.10 Farmers’ willingness to adopt additional BMPs under future climate conditions.

The factors that might influence farmers’ intention to adopt more BMPs were examined

with binary logistic regression (Table 4.4). Concern over the local aquatic environment and

household income are significantly associated with farmers’ willingness of adopting additional

BMPs. Those who are more concerned about water quality issues were more likely to adopt

additional BMPs; for 1 unit increase of environmental concern score, the odds of a farmer to

adopt more BMPs was 1.13 times, when other variables were held constant (Table 4.4).

Farmers of the $25,000 - $49,999/yr income group were significantly less willing to adopt more

BMPs comparing with other income groups. These findings are consistent with that of farmers’

131 interest in water quality trading (WQT) in Chapter 2, where farmers with higher environmental concern were more likely to participate in WQT while the $25,000 - $49,999/yr income group was least likely to participate.

YADD Variables Odds Ratio P

Action1 1.9056 0.4800

Observation 1.1021 0.6578

LakeErie1 6.8205 0.1013

T.Concern 1.1306 0.0405

TillRank2 0.0506 0.0673

TillRank3 0.2346 0.3673

TillRank4 0.3063 0.4863

CoverCrop1 0.7541 0.7552

TotalUsing 1.0181 0.9170

Age 1.0278 0.3706

Gender1 14.9289 0.0567

Education2 0.4383 0.4400

Education3 0.9066 0.9390

Education4 4.6960 0.4830

Live_yr 1.0023 0.8678

Own 0.9959 0.5454

Lease 0.9944 0.4231

Land 1.0043 0.5093

LeaseP 18.0964 0.2957

Manage_yr 0.9775 0.0994

Table 4.4 Binary logistic regression results for farmers’ intention to adopt additional BMPs and associated variables (presented in odds ratio).

132

YADD Variables Odds Ratio P

Income2 0.0408 0.0174

Income3 0.0704 0.2076

Income4 0.0390 0.0657

Income5 0.0671 0.1180

Income_role2 11.5500 0.0675

Income_role3 9.8181 0.0862

Income_role4 4.1457 0.2925

Table 4.4 Continued.

As the farmers’ BMP adoption process is very complicated, the factors associated with BMP

adoption and their influences were found inconsistent among different studies (Knowler and

Bradshaw, 2007; Prokopy et al., 2008; and Liu et al., 2018). It is also the case for additional BMP

adoption under climate change scenario in the Blanchard River watershed. For example,

previous studies (Prokopy, 2008; Lamba et al. 2009; Robinson and Napier 2002, Dasgupta et al.

(2007), Asafu-Adjaye (2008), Paudel et al. (2008), and Wu and Babcock (1998)) found that

education, age and farm size were significantly associated with farmers’ BMP adoption.

However, in this study, these factors did not play a significant role. All farmers who responded to the education level question had at least finished high school, 27% had a college degree and

7% had a graduate degree. There was no significant difference in the willingness to adopt more

BMPs among all the educational level groups. The ratio of leased/owned farm land was also an important factor in other studies. Previous studies posited that farmers who farmed on more leased land were less active in using conservation measures (Xie, 2014; Soule et al. 2000),

133 because they were less motivated to invest in conservations that generally have long-term rather than short-term benefits. Our results did not find a significant association between the percentage of leased land managed by farmers and likelihood of adopting conservation measures.

4.4 Discussion

Weber (2010) argues that climate change could not be easily and accurately observed by lay people, because it is a statistical phenomenon that reflects systematic changes in an average condition for a region. Although climate change seems to be open to everyone to observe and evaluate, the confusion between “climate” and “weather” by many people usually undermines the reliability of personal observation (Weber, 2010). However, Weber (2010) also found that personal observation and experience could be an index of people’s intention to take action. Lack of first-hand experience of climate change’s adverse consequences is one reason that people do not take action (Spencer, 2011). Personal experience captures a person’s attention. People tend to recall the changes in temperature and precipitation that are consistent with their expectation of the local climate change. People who believed that climate change was happening in their region usually claimed that they observed those changes, while those who believed that climate is constant usually claimed that they observed no change, or they considered the changes to be normal, “the climate is always changing”, as one farmer said in our survey. The Corn Belt survey (Loy et al, 2013) found that 66.1% of the farmers in

Western Lake Erie Basin, of which the studied Blanchard River watershed is a subwatershed,

134

believed that climate change was happening (either due to natural or anthropogenic causes or

both), 30.5% thought that there was insufficient evident to judge and only 3.4% did not believe

climate change was happening. This study did not ask directly whether a farmer believed that climate change was happening, but I found 72% of the responded farmers in my survey had at least observed one change in the climate that was consistent with the NCA record; respondents who observed opposite or no change was about 28%. The similar percentage of farmers who believed climate change was happening (66.1%) and that of farmers who had observed changes

(72%) suggest a majority of farmers in this region may have relevant personal experience with climate change that could influence their intention to take action. It might be difficult to reach a consensus on the existence of climate change and its anthropogenic cause in certain groups, but the disagreement does not necessarily prevent these people from taking action to mitigate and/or adapt to climate change. In fact, most people are neither “nonbelievers” who refuse any efforts to mitigate and/or adapt to climate change, nor are they “believers” who promise to take action (O'Connor, 1999). Farmers, for example, are close observers of the climate, although not all of them made accurate observations of climate change trends nor believed

anthropogenic climate change was happening. About 44% of farmers in this study had already taken some action to address the changes they observed, and 77% agreed to take additional action after reading a scientific projection of future changes in their area. Moreover, observation of changes in local climate had significant influence on their action to address those changes. The more accurate observations they made (the changes they observed were consistent with the scientific records, e.g. NCA), the more likely participants were to take action.

More research is needed to understand how personal climate observation influence mitigation

135

and adaption actions, meanwhile, in practice, conservation program should focus on helping farmers enhance the accuracy of personal climate observation by education of climatology and

meteorology knowledge, or by equipping them with simple observation tools such as rain gauges that are low cost.

Personal experience is considered a powerful driver of risk perceptions that will increase the perceived likelihood of a risk if it has recently been experienced (Tversky, 1973). The key is that people are able to causally connect their personal experience or observation to a phenomenon that increases their concern (Weber, 2010). For example, the experience of a flood or a drought makes a person more concerned about climate change because he/she perceives it was an adverse consequence of climate change. In most places, extreme weather events happen only with small probability, thus the connection between these events and climate change is not easily formed. In the Blanchard River watershed however, flooding has been a major concern since the area was developed from a large marsh to farmland and cities over a century ago (Figure 4.6). This might be at least part of the reason why farmers in this area were more concerned about climate and became willing to take actions. In studies of other areas, such as Leiserowitz et al.’s (2008) survey in Florida and the Arctic Climate Impact

Assessment (2004) in Alaska, it was found that the personal experience of people who had been frequently exposed to evidence of climate change had higher willingness to take action; while in Upper Midwest, Morton et al. (2017) found that farmers who had recent experience of flood or drought still had low willingness to change current agricultural practices and strategies, as they perceived that there was too much uncertainty about climate change to justify the actions. The awareness of the connection between local weather events and climate change

136

makes the climate issues more tangible and less distant and people are more likely to feel that

their behaviors are possible to make impactful changes (Spencer, 2011). Knowing that they

might have an impact may encourage people to take action (Weber 2006).

This study found 50% or more of the responded farmers did not observe changes in precipitation, floods or summer temperature, suggesting there was still a significant gap between farmers’ observation of changing climate and the scientific reality. This translated into a lower perception of risk and consequently slower response to implementing conservation

measures to mitigate or adapt to the new environmental challenges that a changing climate will

bring. This is a challenge facing WQT programs if they are to meet the challenge of climate

change. Fundamentally EPA’s TMDL’s will need to consider this as TMDL’s are the driver setting regulations for most programs. It is notable that factors that significantly influenced farmers’ willingness to adopt additional BMPs in a climate change scenario were environmental concern and income, which were also found to be significant in the willingness to participate in WQT.

This finding suggests that enhancing environmental concern and targeting the right income group might be the key to the success of future conservation programs in the Blanchard River watershed. To enhance environmental concern, there will need to be significant changes in farmers’ perceptions of the risk that is presented with the growing number of weather events that will increase nutrient flows and damage crops. As discussed above, it might be a useful strategy to emphasize in conservation and public education programs the connections between climate change and local weather events as well as people’s personal climate observation and experience.

137

4.5 Conclusion

This study found that farmers’ observation of changes in climate was a key driver of their action. Farmers who had already taken action were those who observed climate trends more accurately. However, 50% or more of the farmers did not observe consistent changes in precipitation, floods or summer temperature with scientific reality. This might be a reason that only limited actions were in place to address the impacts of climate change. On the other hand, the personal observation of climate did not significantly affect farmers’ intentions to adopt additional adaptive practices in the future. Similar to that for willingness to participate in WQT, environmental concerns and income were more important for future BMP adoption. To enhance farmers’ environmental awareness on local water quality issues by educating farmers about the local impact of climate change, helping them to understand the causal relationship between their observations and experience and climate change, as well as improving the climate observation accuracy, would benefit BMP promotion via WQT in a climate change context.

In addition, we also see other possibilities for increasing conservation implementation.

Although this study did not include questions about insurance, there has been an increase in the number of acres enrolled in the crop insurance 70% and 69% for corn and soybeans in Ohio, respectively (USDA Risk Management Agency 2017). The number of crop acres insured in Ohio has risen from 4,548,348 acres in 2002 to 6,998, 807 in 2016 which is a 54% increase (USDA Risk

Management Agency 2017). The fact that the gross premiums for that insurance have risen more than four-fold provides evidence that crop losses are rising although the causes could

138

include factors other than climate change. The fact that the US government subsidizes almost

40% of the premiums (Claassen et al. 2017) likely further dampens incentives to change

practices or cropping strategies. Certainly, future research is warranted on the relationship

between risk perception, climate change, and crop insurance participation.

References

Arbuckle, J. G., Prokopy, L. S., Haigh, T., Hobbs, J., Knoot, T., Knutson, C., Loy, A., Mase, A. S.,

McGuire, J., & Morton, L. W. (2013). Climate change beliefs, concerns, and attitudes

toward adaptation and mitigation among farmers in the Midwestern United States.

Climatic Change, 117(4), 943-950.

Arctic Climate Impact Assessment. Impacts of a Warming Arctic. Cambridge, UK: Cambridge

University Press; 2004.

Cavlin, L. (1992). Participation in the U.S. Federal Crop Insurance Program. Commodity

Economics Division, Economic Research Service, U.S. Department of Agriculture. Retrieved

from: https://naldc.nal.usda.gov/download/CAT10879095/PDF

Claassen, R., Langpap, C. and Wu, J. (2017) Impacts of Federal Crop Insurance on Land Use and

Environmental Quality. American Journal of Agricultural Economics, 99 (3): 592-613.

Freudenburg, W. R. (1993). Risk and recreancy: Weber, the division of labor, and the rationality

of risk perceptions. Social Forces, 71(4), 909-932.

139

Gramig, B. M., Barnard, J. M., & Prokopy, L. S. (2013). Farmer beliefs about climate change and

carbon sequestration incentives. Climate Research, 56(2), 157-167.

Hatfield, J., Takle, G., Grotjahn, R., Holden, P., Izaurralde, R.C., Mader, T., Marshall, E., and

Liverman, D. (2014) Agriculture: Climate change impacts in the United States. The Third

National Climate Assessment. Washington, D.C.

Howden, S. M., Soussana, J. F., Tubiello, F. N., Chhetri, N., Dunlop, M., & Meinke, H. (2007).

Adapting agriculture to climate change. Proceedings of the National Academy of Sciences of

the United States of America, 104(50), 19691-19696.

Iowa State University Extension and Outreach. (2014). Iowa nutrient reduction strategy: A

science and technology-based framework to assess and reduce nutrients to Iowa waters

and the Gulf of Mexico. Ames, IA.

Kahan, D. M., Jenkins‐Smith, H., & Braman, D. (2011). Cultural cognition of scientific consensus.

Journal of Risk Research, 14(2), 147-174.

Lamba, P., Filson, G., & Adekunle, B. (2009). Factors affecting the adoption of best management

practices in southern Ontario. The Environmentalist, 29(1), 64.

Leiserowitz A, Broad K. Florida: Public Opinion on Climate Change. A Yale University / University

of Miami / Columbia University Poll. New Haven, CT: Yale Project on Climate Change; 2008,

http://environment.yale.edu/uploads/FloridaGlobalWarmingOpinion.pdf.

140

Loy, A., Hobbs, J., Arbuckle Jr, J., Morton, L., Prokopy, L., Haigh, T., Knoot, T., Knutson, C., Mase,

A., & McGuire, J. (2013). Farmer perspectives on agriculture and weather variability in the

Corn Belt: A statistical atlas. Cscap 0153-2013,

Morton, L. W., Hobbs, J., Arbuckle, J. G., & Loy, A. (2015). Upper Midwest climate variations:

Farmer responses to excess water risks. Journal of Environmental Quality, 44(3), 810-822.

Morton, L. W., Roesch-McNally, G., & Wilke, A. (2017). Upper Midwest farmer perceptions: Too

much uncertainty about impacts of climate change to justify changing current agricultural

practices. Journal of Soil and Water Conservation, 72(3), 215-225.

National Agricultural Statistics Service. (2014). Producers protect or improve millions of acres of

agricultural land. 2012 Census of Agriculture. Washington D.C.

O'Connor, R. E., Bord, R. J., & Fisher, A. (June 01, 1999). Risk Perceptions, General

Environmental Beliefs, and Willingness to Address Climate Change. Risk Analysis, 19, 3,

461-471.

Pryor, S. C., Scavia, D., Downer, C., Gaden, M., Iverson, L., Nordstrom, R., Patz, J., & Robertson,

G. P. (2014). Midwest: Climate change impacts in the United States. The Third National

Climate Assessment. Washington, D.C.

Riley, P. J. and Kiger, G. (2002). Increasing Survey Response: The Drop-Off/Pick-Up Technique.

The Rural Sociologist. 22(1): 6-9.

141

Robinson, J. R., & Napier, T. L. (2002). Adoption of nutrient management techniques to reduce

hypoxia in the gulf of Mexico. Agricultural Systems, 72(3), 197-213.

Roesch-McNally, G. E., Arbuckle, J. G., & Tyndall, J. C. (2017). What would farmers do?

adaptation intentions under a corn belt climate change scenario. Agriculture and Human

Values, 34(2), 333-346.

Spence, A., Poortinga, W., Butler, C., & Pidgeon, N. F. (March 20, 2011). Perceptions of climate

change and willingness to save energy related to flood experience. Nature Climate Change,

1, 1, 46-49.

Tosakana, N., Van Tassell, L., Wulfhorst, J., Boll, J., Mahler, R., Brooks, E., & Kane, S. (2010).

Determinants of the adoption of conservation practices by farmers in the northwest wheat

and range region. Journal of Soil and Water Conservation, 65(6), 404-412.

Tversky, A. & Kahneman, D. Availability: A heuristic for judging frequency and probability.

Cognitive Psychol. 5, 207–232 (1973).

USDA Risk Management Agency. (2017). Ohio Crop Insurance: A Risk Management Agency

State Profile. Washington, D.C. Retrieved from:

https://www.rma.usda.gov/pubs/2017/stateprofiles/ohio16.pdf

Walsh, J., Wuebbles, D., Hayhoe, K., Kossin, J., Kunkel, K., Stephens, G., Thorne, P., Vose, R.,

Wehner, M., Willis, J., Anderson, D., Doney, S., Feely, R., Hennon, P., Kharin, V., Knutson, T.,

Landerer, F., Lenton, T., Kennedy, J., and Somerville, R. (2014). Our changing climate:

142

Climate change impacts in the United States. The Third National Climate Assessment.

Washington, D.C.

Weber, E. U. (2006). Experience-based and description-based perceptions of long-term risk:

Why global warming does not scare us (yet). Climate Change, 77, 103–120.

Weber, E. U. (2010). What shapes perceptions of climate change?. Wiley Interdisciplinary

Reviews: Climate Change, 1(3): 332-342.

Whitmarsh, L. (2008). Are flood victims more concerned about climate change than other

people? The role of direct experience in risk perception and behavioural response. J Risk

Res, 11: 351–374.

Wilson, R., Burnett, L., Ritter, T., Roe, B., & Howard, G. (2013). Farmers, phosphorus and water

quality: A descriptive report of beliefs, attitudes and practices in the Maumee watershed of

northwest Ohio. The Ohio State University, School of Environment & Natural Resources

143

Chapter 5. Conclusion

The studies of this dissertation investigated people’ willingness to adopt conservation

measures with a focus on water quality trading that have the potential to reduce nutrient

runoff to Lake Erie in the context of climate change. A questionnaire survey was conducted

among both farming and non-farming households in the Blanchard River watershed, a

subwatershed of the western Lake Erie basin. For farming households, the survey examined

their current farming practices and their willingness to adopt BMPs in the scenarios of WQT and

climate change; for non-farming households, the willingness to upgrade their household septic

systems as part of a WQT system. The general objective of these studies was to evaluate

whether WQT has the potential to serve as an incentive to promote water quality conservation

BMPs adoption among farmers and septic system upgrades among all rural households.

Chapter 1 was an introduction to the concept of and recent advances in WQT, as well as the

background of this study. Specifically, Chapter 2 aimed to understand how a credit stacking

WQT model affected farmers’ adoption and selection of BMPs, as well as their willingness to participate in WQT; Chapter 3 explored the possibility of expanding the use of WQT from agriculture to household septic system, an often-neglected nonpoint source; Chapter 4 studied farmers’ perception and observation of climate change and how it influenced their actions and willingness to adopt additional BMPs in the future.

For most farmers in the Blanchard River watershed, the All-in-One credit stacking WQT model was preferable to the conventional single credit trading model. Farmers who preferred the All-in-One model were significantly more likely to participate in WQT programs: the odds of

144

being interested in WQT participation was 9.41 times higher than those who preferred models

other than All-in-One, when other variables were held constant. In addition to the preference for the trading model, farmers’ interest in WQT was also related to concerns over local water

quality issues, household income and the role of farming in total household income. Similar

results were also found in the study of farmers’ willingness to adopt additional BMPs to address climate change challenges. Farmers with stronger environmental consciousness and higher

(farming) income were more likely to participate in WQT and to take additional action to mitigate or adapt to climate change. Even for the willingness to upgrade household septic system, people with higher environmental concern scores were significantly more likely to

participate in WQT to upgrade their systems. This finding reflected that people’s participation in

conservation programs were simultaneously influenced by both economic and non-economic incentives. The increased willingness to participate in WQT might be due to the potential of higher economic return, as well as the holistic view of the ecosystem provided by the All-in-One

model. In fact, studies have also found that, as farmers became more environmentally

conscious, they tended to be more motivated by non-economic rather than economic

incentives (Perry-Hill and Prokopy 2014; Ryan et al., 2003 and Greiner et al., 2009). In the

Blanchard River watershed, financial factors such as who receives payment, how and when the

payment would be made, even the price of credit and trading ratio, were given lower priorities

for farmers considering participating in WQT; farmers were most concerned how much

improvement in water quality the WQT program could achieve. This held true from the less

commercial Amish community in Holmes and Wayne County, Ohio (the Alpine Cheese

Phosphorus Trading Plan, Moore, 2014) to in the intensified row crop farms in Hancock County,

145

Ohio studied in this dissertation. However, WQT is often introduced with an emphasis on its economic advantages. This study suggests that non-economic factors, including concerns over local environmental issues, farmers’ role as an environmental steward, trust in program organizers and other stakeholders (Mariola, 2009), etc. are as important as, if not more important than, the economic benefits to the success of a WQT project. Although rarely mentioned in other studies, the ecosystem structure and function integrity represented by the

All-in-One credit stacking model could be a selling point of WQT in addition to its economic

advantages. This finding also implies that, if the ecosystem services market is to be used as a

tool to motivate people conserving nature, focusing only on the market mechanism and

treating ecosystem services as conventional commodities might undermine its effectiveness,

especially when it comes to the cultural, aesthetic, religious, and social services that are uneasy

to quantify with monetary values.

The selection and adoption of specific BMPs is a complicated process. This process

involved a wide range of variables that were found having inconsistent influences on different

BMPs, in different areas and on different BMP adoption stages (Knowler and Bradshaw, 2007;

Prokopy et al., 2008; and Liu et al., 2018). This study found that farmers were most interested in adopting cover crops in the WQT scenario and in the climate change scenario, while less interested in no-till. The adoption of cover crops was related to the previous experience of using this BMP, while the adoption of no-till is influenced more by income and leased land percentage. The difference might due to the different stages adoption. According to the 2012

Agriculture Census (USDA, 2012), about 39.8% of croplands in Ohio was in no-till and 3.3% was

in cover crops. By 2012, there were 20,712 farms had adopted no-till with an average no-till

146 acreage of 207 per farm; and 6,565 farms had adopted cover crops with an average cover crop acreage of 54 per farm (USDA, 2012).

The heterogeneity and complexity of BMP adoption is a challenge for WQT, especially for large-scale programs, therefore, studying the adoption of BMPs on a local scale is more meaningful and practical. In fact, limiting the trading program to a small, bilateral scale could increase participation in WQT (Woodward et al, 2002); small-scale community-based WQT programs also provide more benefits over larger-scaled programs (Moore, 2014). Moreover, conservation programs that are designed according to local conditions are more likely to be supported by the local community. The study of the WQT for septic system upgrades identified a hot spot, which was about the size of two townships within Hancock County, for potential pilot project based on the spatial concentration of households that were most interested in the

WQT idea and most environmentally conscious.

Farmers’ observation of changes in local climate could also motivate actions to adapt to or mitigate climate change challenges. This study found that the more accurate observations they made (the changes they observed being consistent with the scientific records, e.g. NCA), the more likely they would take action. More importantly, people who perceived higher risk would take action if they could make the causal connection between their personal observation or experience and global climate change; in other words, moreover, this happened when they were informed about the local impacts of climate change. Usually, having had an adverse experience with climate events could help people establish the causal connection. In the

Blanchard River watershed, flooding has been a major concern since the area was transformed

147

from the former Great Black Swamp into farmland and cities over a century ago. As heavy rain events become more common in the area, the frequency and intensity of floods will increase,

leading to an enhanced perception of risk and likely interest in WQT. It should be also noted that, it is sometimes impossible and unnecessary to reach a consensus on the existence of climate change and its anthropogenic cause in certain groups, because most people are neither

“nonbelievers” who refuse any efforts to mitigate and/or adapt to climate change, nor are they

“believers” who promise to take action (O'Connor, 1999). Forming a common ground on which different stakeholders can collaborate is important for WQT programs.

This study sheds lights on the pilot WQT study for the Maumee River watershed. Based on the findings of this dissertation research, I recommend that WQT program be introduced to the community while emphasizing both the economic and non-economic benefits of the trading program, with a focus on the environmental issues that concern the farmers. For this purpose, the All-in-One credit stacking model could be an appropriate design. The Maumee River

watershed is a large watershed, so rather than have a WQT for the whole watershed; I suggest that a pilot project be started on a smaller scale, such as one county or the headwater area of a subwatershed. The pilot project might first approach the farming households with higher environmental consciousness and higher farming income. If possible, a survey should be conducted in advance to identify the farms that are most supportive of the idea and perhaps a hot spot where these farms concentrate. Meanwhile, education events such as field days and workshops could be organized to inform farmers about local water quality issues, climate

change impacts on local farms, and local farmer experience using cover crops and other BMPs.

In general, the pilot project should be localized and able to meet the needs of local farmers.

148

For future research, I am interested in exploring the water quantity issues regarding

WQT in addition to the quality issues. In the Blanchard River watershed, the intensity and

frequency of floods have been increasing due to climate change. The city of Findlay, the most influenced area, has been seeking solutions to ameliorate the flooding problem. It typifies a rising problem for all water quality trading programs: the fact that major weather events are

associated with high pollutant discharge and that climate change is causing more severe

weather events--both flooding and droughts. Water quantity and water quality are interrelated

but almost all WQT programs have focused more on quality than quantity. This is mainly

because the TMDL's set NPDES permit limits on point sources within a watershed and the

trading programs target these limits through conservation measures applied throughout the

watershed. However, a pollutant load (in kg) is calculated as Flow × Concentration so both

quantity and quality are involved. The issue of water quantity in trading programs has focused

almost solely on how a point source such as a wastewater treatment plant can maintain its

regulated discharge (quantity) limit through improving its flow (quality). It can do this in a static

mode (same flow and concentration) or by increasing its pipe outflow while decreasing the

concentration of the pollutant. While this is a valuable feature of the NPDES permit system that

enables factories or wastewater treatment plants to increase their productive capacity while

lowering their concentration through facility upgrades or water quality trading, there has not

been the same emphasis on calculating the flow of the stream, especially rain events, relating

to the effectiveness of the conservation measures implemented by the sellers of credit.

In fact, it can be said that every trading program to date has experienced water quantity

issues whether it be floods or droughts that affected the supplier side of credits. However, this

149

is seldom factored into the trading program. These variations in water quantity such as floods

directly affect the effectiveness of conservation measures. EPA has not included climate change

into its TMDL policies. Trading programs in the future would be well-advised to provide more

incentives to directly deal with the forecasted increased in major weather events as outlined in

the fourth chapter of this dissertation. A high percentage of phosphorus and nitrogen runoff

actually happens during the main rain events (Owens et al., 2010; Winslow, 2016), but trading

plans too often calculate baselines during low flow. This has become a major concern in the

U.S., as the areas that are affected by heavy precipitation events have increased in the last century (Walsh et al., 2014). The same trend is also found in the Midwest, according to The

Third Climate Science Special Report of the Fourth National Climate Assessment (Wuebbles et

al., 2017), this trend is expected to continue in the future and will result in flooding, soil erosion,

water quality degradation, and other negative impacts on agriculture, public health,

transportation, and infrastructure.

In the survey for this dissertation, I collected some qualitative data on farmers’ opinions

on solving the flooding problems in the Blanchard River watershed. The farmers were less

impacted by flooding as most of them did not live in the City of Findlay, but as a group, they

would be greatly affected by the proposed solution as the proposed diversion channel and storage basin, which was to be built mostly on cropland. The farmers thought that their interests were largely neglected in the alleviation plan. In particular, the Army Corp of

Engineers’ benefit/cost analysis did not factor in the value of lost crop production or the devaluation of farmlands. Therefore, most farmers strongly opposed the proposed solutions, and there was tension existing between the farming community and the city of Findlay. If a

150

WQT program aims to address water quality and quantity issue, it has to address farmers’

perceptions of the cause of flooding and their attitudes towards different solutions, as farmers

are the potential credit suppliers. How the role of farmers differs between water quality trading

and water quantity trading is worth studying, especially when WQT is to address the impacts of

climate change in the future.

Reference

Greiner, R., Patterson, L., Miller, O. (2009). Motivations, risk perceptions and adoption of

conservation practices by farmers. Agr Syst 99, 86-104.

Knowler, D. & Bradshaw, B. (2007). Farmers’ adoption of conservation agriculture: A review and

synthesis of recent research. Food Policy, 32, 25-48.

Liu, T., Bruins, R. J., & Heberling, M. T. (2018). Factors Influencing Farmers’ Adoption of Best

Management Practices: A Review and Synthesis. Sustainability, 10(2), 432.

Mariola, M. J. (2009). Are Markets the Solution to Water Pollution? A Sociological Investigation

of Water Quality Trading. The Ohio State University.

Moore, R. H., (2014). The role of trading in achieving water quality objectives: Congress

testimony before the Water Resources and Environment Subcommittee Committee on

Transportation and Infrastructure United States House of Representatives. Washington D.C.

151

O'Connor, R. E., Bord, R. J., & Fisher, A. (June 01, 1999). Risk Perceptions, General

Environmental Beliefs, and Willingness to Address Climate Change. Risk Analysis, 19, 3,

461-471.

Owens, L., Bonta, J., & Shipitalo, M. (2010). USDA-ARS north Appalachian experimental

watershed: 70-year hydrologic, soil erosion, and water quality database. Soil Science

Society of America Journal, 74(2), 619.

Perry-Hill, R. & Prokopy, L. (2014). Comparing different types of rural landowners: Implications

for conservation practice adoption. J Soil Water Conserv 69, 266-278.

Prokopy, L. S., Floress, K., Baumgart-Getz, A., & Klotthor-Weinkauf, D. (2008). Determinants of

agricultural best management practice adoption: Evidence from the literature. Journal of

Soil and Water Conservation, 63, 5, 300-311.

Ryan, R.L., Erickson, D.L., De Young, R. (2003). Farmers' motivations for adopting conservation

practices along 801 riparian zones in a mid-western agricultural watershed. J Environ

Planning Manage 46, 19-37.

U.S. Department of Agriculture (USDA). (2012). Census of agriculture. National Agricultural

Statistics Service. Retrieved from:

https://www.agcensus.usda.gov/Publications/2012/index.php.

Walsh, J., Wuebbles, D., Hayhoe, K., Kossin, J., Kunkel, K., Stephens, G., Thorne, P., Vose, R.,

Wehner, M., Willis, J., Anderson, D., Doney, S., Feely, R., Hennon, P., Kharin, V., Knutson, T.,

152

Landerer, F., Lenton, T., Kennedy, J., and Somerville, R. (2014). Our changing climate:

Climate change impacts in the United States. The Third National Climate Assessment.

Washington, D.C.

Winslow, C.J. (2016). Current Research Efforts on Nutrient Load Reduction Methods. Region 5

Harmful Algal Bloom Clean Water Act and Safe Drinking Water Act Workshop.

Woodward, R. T., Kaiser, R. A., & Wicks, A. B. (2002). The structure and practice of water quality

trading markets. JAWRA Journal of the American Water Resources Association, 38(4), 967-

979.

Wuebbles, D.J., D.R. Easterling, K. Hayhoe, T. Knutson, R.E. Kopp, J.P. Kossin, K.E. Kunkel, A.N.

LeGrande, C. Mears, W.V. Sweet, P.C. Taylor, R.S. Vose, and M.F. Wehner, 2017: Our

globally changing climate. In Climate Science Special Report: Fourth National Climate

Assessment, Volume I. D.J. Wuebbles, D.W. Fahey, K.A. Hibbard, D.J. Dokken, B.C. Stewart,

and T.K. Maycock, Eds. U.S. Global Change Research Program, pp. 35-72,

doi:10.7930/J08S4N35.

153

Bibliography

Abdalla, C., Borisova, T., Parker, D., & Blunk, K. S. (2007). Water quality credit trading and

agriculture: Recognizing the challenges and policy issues ahead. Choices, 22(2), 117-124.

Allred, S. B., & Ross-Davis, A. (2011). The drop-off and pick-up method: An approach to reduce

nonresponse bias in natural resource surveys. Small-Scale Forestry, 10(3), 305-318.

Andrews, A.C., Clawson, R.A., Gramig, B.M., Raymond, L. (2013). Why do farmers adopt

conservation tillage? An experimental investigation of framing effects. J Soil Water Conserv

68, 501-511.

Arbuckle, J. G., Prokopy, L. S., Haigh, T., Hobbs, J., Knoot, T., Knutson, C., Loy, A., Mase, A. S.,

McGuire, J., & Morton, L. W. (2013). Climate change beliefs, concerns, and attitudes

toward adaptation and mitigation among farmers in the midwestern united states. Climatic

Change, 117(4), 943-950.

Arctic Climate Impact Assessment. Impacts of a Warming Arctic. Cambridge, UK: Cambridge

University Press; 2004.

Arend, K. K., Beletsky, D., DePINTO, J. V., Ludsin, S. A., Roberts, J. J., Rucinski, D. K., Scavia, D.,

Schwab, D. J., & Höök, T. O. (2011). Seasonal and interannual effects of hypoxia on fish

habitat quality in central lake erie. Freshwater Biology, 56(2), 366-383.

154

Borchardt, M. A., Chyou, P. H., DeVries, E. O., & Belongia, E. A. (2003). Septic system density

and infectious diarrhea in a defined population of children. Environmental Health

Perspectives, 111(5), 742-748.

Bosch, N. S., Evans, M. A., Scavia, D., & Allan, J. D. (2014). Interacting effects of climate change

and agricultural BMPs on nutrient runoff entering lake erie. Journal of Great Lakes

Research, 40(3), 581-589.

Brandt, B. & Baird J. (2008). BMP Challenge: Yield and Income Risk Protection for Corn Farmers

Who Adopt Water Quality BMPs. Presentation. Retrieved from:

http://www.dep.state.pa.us/dep/subject/advcoun/ag/2008/August2008/BMP%20Challeng

e%20PA%20.pdf

Braskerud, B., Hartnik, T., & Løvstad, Ø. (2005). The effect of the redox-potential on the

retention of phosphorus in a small constructed wetland. Water Science and Technology,

51(3-4), 127-134.

Breetz, H. L., Fisher-Vanden, K., Jacobs, H., & Schary, C. (2005). Trust and communication:

Mechanisms for increasing farmers’ participation in water quality trading. Land Economics,

81(2), 170-190.

Carroll, N., & Jenkins, M. (2013). The matrix: Mapping ecosystem service markets. Ecosystem

Marketplace. Retrieved from: http://www.ecosystemmarketplace.com/wp-

content/uploads/2015/09/the_matrix.pdf

155

Cavlin, L. (1992). Participation in the U.S. Federal Crop Insurance Program. Commodity

Economics Division, Economic Research Service, U.S. Department of Agriculture. Retrieved

from: https://naldc.nal.usda.gov/download/CAT10879095/PDF

Chouinard, H.H., Wandschneider, P.R., Paterson, T. (2016). Inferences from sparse data: An

integrated, meta-630 utility approach to conservation research. Ecol Econ 122, 71-78.

Claassen, R., Langpap, C. and Wu, J. (2017). Impacts of Federal Crop Insurance on Land Use and

Environmental Quality. American Journal of Agricultural Economics, 99 (3): 592-613.

Cooley, D., & Olander, L. (2011). Stacking ecosystem services payments: Risks and solutions.

Nicholas Institute for Environmental Policy Solutions, Working Paper NI WP, , 11-04.

Corrales, J., Naja, G. M., Bhat, M. G., & Miralles-Wilhelm, F. (2014). Modeling a phosphorus

credit trading program in an agricultural watershed. Journal of Environmental

Management, 143(0), 162-172.

Costanza, R., d’Arge, R., De Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K., Naeem, S.,

O’Neill, R. V., & Paruelo, J. (1998). The value of ecosystem services: Putting the issues in

perspective. Ecological Economics, 25(1), 67-72.

Costanza, R., de Groot, R., Sutton, P., van der Ploeg, S., Anderson, S. J., Kubiszewski, I., Farber, S.,

& Turner, R. K. (2014). Changes in the global value of ecosystem services. Global

Environmental Change, 26, 152-158.

156

Cousino, L. K., Becker, R. H., & Zmijewski, K. A. (2015). Modeling the effects of climate change

on water, sediment, and nutrient yields from the Maumee river watershed

doi:https://doi.org/10.1016/j.ejrh.2015.06.017

Daily, G. (1997). Nature's services: Societal dependence on natural ecosystems Island Press.

Damon, M. (2017). Current Agricultural Use Value. Members Only. 132(7), 1-10. Retrieved from:

https://www.lsc.ohio.gov/documents/reference/current/membersonlybriefs/132cauv.pdf

Dobiesz, N. E., Hecky, R. E., Johnson, T. B., Sarvala, J., Dettmers, J. M., Lehtiniemi, M., Rudstam,

L. G., Madenjian, C. P., & Witte, F. (2010). Metrics of ecosystem status for large aquatic

systems—a global comparison. Journal of Great Lakes Research, 36(1), 123-138.

Ecosystem Marketplace Initiative. (2016). Alliances for green infrastructure: state of watershed

investment 2016. Forest Trends. Retrieved from: http://www.forest-

trends.org/documents/files/doc_5463.pdf

Electric Power Research Institute (EPRI). (2014). Ohio River Basin Water Quality Trading Project.

Retrieved from: http://wqt.epri.com/pdf/3002001739_WQT-Program-Summary_2014-

03.pdf

Emerson, J. W. (1971). Channelization: A case study. Science (New York, N.Y.), 173(3994), 325-

326.

Eugene, P. O. (1962). Relationships between structure and function in the ecosystem. Japanese

Journal of Ecology, 12(3), 108-118.

157

Fang, F., Easter, K. W., & Brezonik, P. L. (2005). Point‐Nonpoint source water quality trading: A

case study in the Minnesota river basin. JAWRA Journal of the American Water Resources

Association, 41(3), 645-657.

Fleming, P., Lichtenberg, E., Newburn, D.A. (2015). Agricultural Cost Sharing and Conservation

Practices for Nutrient Reduction in the Chesapeake Bay Watershed, 2015 AAEA & WAEA

Joint Annual Meeting, July 26-28, San Francisco, California. Agricultural and Applied

Economics Association & Western Agricultural Economics Association.

Fox, J. (2008). Getting two for one: Opportunities and challenges in credit stacking.

Conservation and Biodiversity Banking: A Guide to Setting Up and Running Biodiversity

Credit Trading Systems.

Freudenburg, W. R. (1993). Risk and recreancy: Weber, the division of labor, and the rationality

of risk perceptions. Social Forces, 71(4), 909-932.

Gao, Y. (2016). China's response to climate change issues after paris climate change conference.

Advances in Climate Change Research, 7(4), 235-240.

García, J. H., Heberling, M. T., & Thurston, H. W. (2011). Optimal pollution trading without

pollution reductions: A note. Journal of the American Water Resources Association, 47(1),

52-58.

Gedikoglu, H., McCann, L.M. (2012). Adoption of win-win, environment-oriented, and profit-

oriented practices among livestock farmers. J Soil Water Conserv 67, 218-227.

158

Gramig, B. M., Barnard, J. M., & Prokopy, L. S. (2013). Farmer beliefs about climate change and

carbon sequestration incentives. Climate Research, 56(2), 157-167.

Green, D., & Shapiro, I. (1996). Pathologies of rational choice theory: A critique of applications

in political science. Yale University Press.

Green, J. E. (2002). Evaluating phosphorus migration from septic systems near Otsego Lake.

34th Ann.Rept.(2001).SUNY Oneonta Biol.Fld.Sta., SUNY Oneonta.

Greiner, R., Patterson, L., Miller, O. (2009). Motivations, risk perceptions and adoption of

conservation practices by farmers. Agr Syst 99, 86-104.

Haghjou, M., Hayati, B., Choleki, D.M. (2014). Identification of Factors Affecting Adoption of Soil

Conservation 664 Practices by Some Rainfed Farmers in Iran. J Agric Sci Technol 16, 957-

967.

Hatfield, J., Takle, G., Grotjahn, R., Holden, P., Izaurralde, R.C., Mader, T., Marshall, E., and

Liverman, D. (2014) Agriculture: Climate change impacts in the United States. The Third

National Climate Assessment. Washington, D.C.

Heberling, M. T. (2011). Issues in water quality trading: Introduction to featured collection.

Journal of the American Water Resources Association, 47(1), 1-4.

Herrman, K. S. (2007). Mechanisms controlling nitrogen removal in agricultural headwater

streams. The Ohio State University.

159

Hess, M. (2004). Spatial relationships towards a reconceptualization of embeddedness.

Progress in Human Geography, 28(2), 165-186.

Ho, J. C., & Michalak, A. M. (2017). Phytoplankton blooms in Lake Erie impacted by both long-

term and springtime phosphorus loading. Journal of Great Lakes Research, 43(3), 221-228.

Hodaj, A., Bowling, L. C., Frankenberger, J. R., & Chaubey, I. (2017). Impact of a two-stage ditch

on channel water quality. Agricultural Water Management, 192, 126-137.

Horan, R. D., & Shortle, J. S. (2011). Economic and ecological rules for water quality trading.

Journal of the American Water Resources Association, 47(1), 59-69.

Howden, S. M., Soussana, J. F., Tubiello, F. N., Chhetri, N., Dunlop, M., & Meinke, H. (2007).

Adapting agriculture to climate change. Proceedings of the National Academy of Sciences of

the United States of America, 104(50), 19691-19696.

Iowa State University Extension and Outreach. (2014). Iowa nutrient reduction strategy: A

science and technology-based framework to assess and reduce nutrients to Iowa waters

and the Gulf of Mexico. Ames, IA.

Jones, C., McGee, B., Epstein, L., Fisher, E., Sanner, P., & Gray, E. Nutrient trading by municipal

stormwater programs in Maryland and Virginia: Three case studies.

Kahan, D. M., Jenkins‐Smith, H., & Braman, D. (2011). Cultural cognition of scientific consensus.

Journal of Risk Research, 14(2), 147-174.

160

Kalcic, M. M., Kirchhoff, C., Bosch, N., Muenich, R. L., Murray, M., Griffith Gardner, J., & Scavia,

D. (2016). Engaging stakeholders to define feasible and desirable agricultural conservation

in western lake erie watersheds. Environmental Science & Technology, 50(15), 8135-8145.

Kane, D. D., Conroy, J. D., Richards, R. P., Baker, D. B., & Culver, D. A. (2014). Re-eutrophication

of Lake Erie: Correlations between tributary nutrient loads and phytoplankton biomass.

Journal of Great Lakes Research, 40(3), 496-501.

King, D. M. (2005). Crunch time for water quality trading. Choices, 20(1), 71-75.

King, D. M., & Kuch, P. J. (2003). Will nutrient credit trading ever work? an assessment of supply

and demand problems and institutional obstacles. Environmental Law Reporter News and

Analysis, 33(5), 10352-10368.

Knowler, D. & Bradshaw, B. (2007). Farmers’ adoption of conservation agriculture: A review and

synthesis of recent research. Food Policy, 32, 25-48.

Kramer, J. (2003). Lessons from the trading pilots: Applications for Wisconsin water quality

trading policy. Resource Strategies, Inc., Madison, WI

Lamba, P., Filson, G., & Adekunle, B. (2009). Factors affecting the adoption of best management

practices in southern Ontario. The Environmentalist, 29(1), 64.

Leiserowitz A, Broad K. Florida: Public Opinion on Climate Change. A Yale University / University

of Miami / Columbia University Poll. New Haven, CT: Yale Project on Climate Change; 2008,

http://environment.yale.edu/uploads/FloridaGlobalWarmingOpinion.pdf.

161

Levy, S. (2017). Learning to love the great black swamp. UNDark. Retrieved from:

https://undark.org/article/great-black-swamp-ohio-toledo/

Ligon, F.K., Dietrich, W.E., Trush, W.J., (1995). Downstream ecological effects of dams.

BioScience, 45, 183–192.

Lindner, J. R., Murphy, T. H., & Briers, G. E. (2001). Handling nonresponse in social science

research. Journal of Agricultural Education, 42(4), 43-53.

Liu, T., Bruins, R. J., & Heberling, M. T. (2018). Factors Influencing Farmers’ Adoption of Best

Management Practices: A Review and Synthesis. Sustainability, 10(2), 432.

Loy, A., Hobbs, J., Arbuckle Jr, J., Morton, L., Prokopy, L., Haigh, T., Knoot, T., Knutson, C., Mase,

A., & McGuire, J. (2013). Farmer perspectives on agriculture and weather variability in the

Corn Belt: A statistical atlas. Cscap 0153-2013.

Macintosh, K., Jordan, P., Cassidy, R., Arnscheidt, J., & Ward, C. (2011). Low flow water quality

in rivers; septic tank systems and high-resolution phosphorus signals. Science of the Total

Environment, 412, 58-65.

Mancl, K., & Slater. B. (2013). Suitable of Ohio soils for treating wastewater. OSU Extension

Bulletin.

Mariola, M. J. (2009). Are Markets the Solution to Water Pollution? A Sociological Investigation

of Water Quality Trading. The Ohio State University.

162

Marshall, E. P., & Weinberg, M. (2012). Baselines in Environmental Markets: Tradeoffs between

Cost and Additionality. United States Department of Agriculture, Economic Research

Service.

Maryland Water Quality Trading Advisory Committee. (2017). Maryland trading and offset

policy and guidance manual Chesapeake Bay watershed. Maryland Department of

Environment. Retrieved

from:http://mde.maryland.gov/programs/water/Documents/WQTAC/TradingManualUpda

te4.17.17.pdf

Maxted, J. R., Barbour, M. T., Gerritsen, J., Poretti, V., Primrose, N., Silvia, A., Penrose, D. &

Renfrow, R. (2000). Assessment framework for mid-Atlantic coastal plain streams using

benthic macroinvertebrates. Journal of the North American Benthological Society.19: 128–

144.

McDonald, J. H. (2009). Handbook of biological statistics Sparky House Publishing Baltimore,

MD.

McSweeny, W. T., & Kramer, R. A. (1986). The integration of farm programs for achieving soil

conservation and nonpoint pollution control objectives. Land Economics, 62(2), 159-173.

Meehan, H. (2004). Phosphorus migration from a near-lake septic system in the Otsego Lake

watershed, summer 2003. 36th Annual Report (2003).SUNY Oneonta Bio.Fld.Sta., SUNY

Oneonta.

163