Direct and Indirect Ecological Consequences of Human
Activities in Urban and Native Ecosystems
by
Elizabeth Miessner Cook
A Dissertation Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy
Approved June 2014 by the Graduate Supervisory Committee:
Sharon J. Hall (Chair) Christopher G. Boone Scott L. Collins Nancy B. Grimm
ARIZONA STATE UNIVERSITY
August 2014 ABSTRACT
Though cities occupy only a small percentage of Earth’s terrestrial surface, humans concentrated in urban areas impact ecosystems at local, regional and global scales. I examined the direct and indirect ecological outcomes of human activities on both managed landscapes and protected native ecosystems in and around cities. First, I used highly managed residential yards, which compose nearly half of the heterogeneous urban land area, as a model system to examine the ecological effects of people’s management choices and the social drivers of those decisions. I found that a complex set of individual and institutional social characteristics drives people’s decisions, which in turn affect ecological structure and function across scales from yards to cities. This work demonstrates the link between individuals’ decision-making and ecosystem service provisioning in highly managed urban ecosystems.
Second, I examined the distribution of urban-generated air pollutants and their complex ecological outcomes in protected native ecosystems. Atmospheric carbon dioxide (CO2), reactive nitrogen (N), and ozone (O3) are elevated near human activities and act as both resources and stressors to primary producers, but little is known about their co-occurring distribution or combined impacts on ecosystems. I investigated the urban “ecological airshed,” including the spatial and temporal extent of N deposition, as well as CO2 and O3 concentrations in native preserves in Phoenix, Arizona and the outlying Sonoran Desert. I found elevated concentrations of ecologically relevant pollutants co-occur in both urban and remote native lands at levels that are likely to affect ecosystem structure and function. Finally, I tested the combined effects of CO2, N, and
i O3 on the dominant native and non-native herbaceous desert species in a multi-factor dose-response greenhouse experiment. Under current and predicted future air quality conditions, the non-native species (Schismus arabicus) had net positive growth despite physiological stress under high O3 concentrations. In contrast, the native species
(Pectocarya recurvata) was more sensitive to O3 and, unlike the non-native species, did not benefit from the protective role of CO2. These results highlight the vulnerability of native ecosystems to current and future air pollution over the long term. Together, my research provides empirical evidence for future policies addressing multiple stressors in urban managed and native landscapes.
ii ACKNOWLEDGMENTS
Foremost, I would like to thank Sharon Hall for her endless support, training, and guidance toward becoming a scientist and broad critical thinker. Thank you! I also thank my committee, Christopher Boone, Scott Collins, and Nancy Grimm, for your encouragement and support in pursing an interdisciplinary graduate education and my research goals, as well as for your helpful feedback improving this research.
Thank you to the many members of the IGERT in Urban Ecology community, the School of Life Sciences graduate group and especially past and current Hall lab members, the many Grimm Lab reading groups, and the Central Arizona-Phoenix Long-term
Ecological Research network. Most especially, thank you to Stevan Earl, Rebecca Hale,
David Huber, David Iwaniec, Kelli Larson, Jennifer Learned, Yevgeniy Marusenko,
Pamela Padgett, and Katelyn Parady. Thank you for many thoughtful, engaging, and inspiring conversations and collaborations that have shaped my perspective on science and research. I am lucky to have been a part of these collaborative and engaging communities.
To my family and parents, thank you for teaching (and reminding) me that there are no limits to one’s dreams or capabilities and for supporting me in my many endeavors – that is the greatest gift. Finally, thank you to my love, Lyzzette Foz Bullock, for your innumerable measures of support and for your unwavering belief in me. I could not have done this without you.
iii TABLE OF CONTENTS
Page
LIST OF TABLES ...... ix
LIST OF FIGURES ...... xi
CHAPTER
1 INTRODUCTION ...... 1
Scope and Structure of this Dissertation ...... 2
References ...... 7
2 RESIDENTIAL LANDSCAPES AS SOCIAL-ECOLOGICAL SYSTEMS:
SYNTHESIS OF MULTI-SCALAR INTERACTIONS BETWEEN PEOPLE
AND THEIR HOME ENVIRONMENT ...... 9
Abstract ...... 9
Introduction ...... 11
Methods ...... 13
Findings ...... 15
Ecology of Residential Landscapes ...... 16
Management Decisions ...... 29
Multi-Scalar Human Drivers of Residential Landscapes ...... 32
Legacy Effects ...... 42
Discussion and Future Research Directions ...... 44
Conclusions ...... 49
Acknolwedgements ...... 51
References ...... 52
iv CHAPTER Page
3 ATMOSPHERIC NITROGEN DEPOSITION IN A LARGE SEMI-ARID CITY
IS LOWER THAN EXPECTED: FINDINGS FROM A METHODS
COMPARISON ...... 80
Abstract ...... 80
Introduction ...... 82
Methods ...... 87
Site Characteristics ...... 87
Long-Term N Deposition Monitoring ...... 88
Estimating Dry Deposition with Multiple Sampling Approaches ...... 91
Meteorological and Urban Site Characteristics ...... 96
Data Analyses ...... 98
Results ...... 99
Long-Term N Deposition ...... 99
Gaseous N Concentrations and Dry Deposition ...... 104
Total N Deposition Comparison ...... 106
Discussion ...... 108
Patterns of Nitrogen Deposition in a Large Semi-Arid City and Surrounding
Native Ecosystem ...... 108
Drivers and Sources of Nitrogen Deposition in an Arid Metropolitan
Region ...... 112
Underestimation of N Deposition in Arid Systems ...... 116
Uncertainties in Urban Arid N Deposition Estimates ...... 117
v CHAPTER Page
Conclusion ...... 120
Acknolwedgements ...... 122
References ...... 123
4 CHARACTERIZATION OF THE “ECOLOGICAL AIRSHED” OF NEAR-
URBAN PROTECTED ECOSYSTEMS ...... 145
Abstract ...... 145
Introduction ...... 147
Methods ...... 151
Experimental Design and Site Characteristics ...... 151
O3 and N Field Collection and Analyses ...... 153
Calculating O3 and N Concentrations and Dry N Deposition ...... 155
Monitoring CO2 Concentrations ...... 157
Data Analyses ...... 159
Results ...... 160
Ozone ...... 160
Nitric Acid ...... 161
Ammonia ...... 161
Nitrogen Oxide ...... 162
Total Dry N Deposition ...... 162
Carbon Dioxide Concentrations ...... 164
Edge-to-Interior Open Space Transect ...... 165
Discussion ...... 166 vi CHAPTER Page
Spatial and Temporal Co-Occurrence of CO2, O3, and N in an Urban
Ecological Airshed ...... 166
Summer Regional O3 Patterns ...... 168
Regional N Concentrations and Deposition ...... 169
Regional and Temporal CO2 Concentrations ...... 171
Urban “Ecological Airshed” Extends to Interior of an Urban Open Space
Preserve ...... 175
Conclusion ...... 176
Acknowledgements ...... 177
References ...... 178
5 SYNERGISTIC NET EFFECTS OF URBAN AIR POLLUTION ON
ECOSYSTEMS ...... 201
Abstract ...... 201
Introduction ...... 203
Methods ...... 208
Experimental Design ...... 208
Study Species ...... 209
Experimental Treatment Set-up ...... 210
Response Variables ...... 212
Data Analyses ...... 214
Results ...... 216
Non-Native Species Responses under Single and Combined Factors ...... 216 vii CHAPTER Page
Native Species Affected Most by O3 ...... 218
Additive and Non-Additive Effects of Combined Factors ...... 219
Discussion ...... 222
Responses to Co-Occurring Pollutants ...... 223
Predicted Impacts of Water Co-Limitation ...... 226
Synergistic Responses to Co-Occurring Polltuants ...... 228
Implications of Urban Air Quality for Primary Producers and
Ecosystems ...... 229
Conclusion ...... 231
Acknowledgements ...... 232
References ...... 233
6 CONCLUDING REMARKS ...... 250
Synthesis of Major Research Findings ...... 250
Next Steps and Implications for Future Management ...... 252
References ...... 255
REFERENCES...... 257
viii LIST OF TABLES
Table Page
2.1 Variables Related to Ecological Properties of Residential Landscapes ...... 73
2.2 Variables Related to Ecological Functioning and Ecosystem Services of
Residential Landscapes ...... 74
2.3 Variables Related to Residential Landscape Management Decisions ...... 75
2.4 Variables Related to Multi-Scalar Human Drivers of Residents’ Preferences and
Management Practices ...... 76
2.5 Future Research Opportunities ...... 77
3.1 Site Characteristics of N Deposition Monitoring ...... 130
3.2 N Deposition Velocities for Arid, Semi-Arid and Urban Ecosystems ...... 131
3.3 Throughfall, Bulk, and Dry N Deposition across Sites ...... 132
3.4 Predicting N Deposition from Precipitation Variables ...... 133
3.5 Predicting N Deposition from Precipitation and Site Characteristics ...... 134
3.6 Total N Deposition Estimates from Multiple Methods and Percent Difference in N
Deposition Estimates ...... 135
4.1 Site Characteristics within Open Space Preserves in Phoenix, Arizona ...... 186
-3 4.2 Mean Site O3, HNO3, NH3, NOx and Total N Concentrations (µg m ) ...... 187
-1 -1 4.3 Mean Site HNO3, NH3, NOx and Total N Deposition (kg N ha yr ) ...... 188
-1 -1 4.4 Ozone, HNO3, NH3, NOx and Total N Deposition (kg N ha yr ) along Edge-to-
Interior Transect of Protected Urban Open Space ...... 189
5.1 Carbon Dioxide and O3 Concentrations in Treatment Chambers ...... 242
ix Table Page
5.2 Linear Mixed Model Results Showing Individual, Additive or Interactive Effects on
Growth Rate and Photosynthetic Yield ...... 243
x LIST OF FIGURES
Figure Page
2.1 Published Literature on Social and Ecological Aspects of Residential
Landscapes ...... 78
2.2 Conceptual Framework of Multi-Scalar Social-Ecological Interaction Landscapes
of Residential Landscapes ...... 79
3.1 Long-Term Nitrogen Deposition Monitoring Sites ...... 136
3.2 Long-Term Seasonal Average of N Throughfall Deposition and Precipitation .... 137
3.3 Mean Ambient HNO3, NH3, and NOx Concentrations at Urban and Outlying
Locations between 2010 - 2012 ...... 138
3.4 Percent HNO3, NH3, and NOx of Total Dry N Deposition ...... 139
3.5 Urban and Outlying Dry N Deposition Estimated by the Inferential Method ...... 140
3.6 Predicted Total N Deposition across Multiple Seasons ...... 141
4.1 Nitrogen, O3 and CO2 Monitoring Sites in Metropotian Region of Phoenix, AZ . 190
4.2 Field Set-up for Measuring Ambient Atmospheric Gaseous Concentrations ...... 191
4.3 N Deposition, O3 and CO2 in Outlying and Urban Locations over 5 Summer
Sampling Intervals ...... 192
4.4 Mean Summer N Deposition, O3 and CO2 in Outlying and Urban Locations ...... 193
4.5 Summer Hourly Urban and Outlying CO2 Concentrations at 2 m above Ground . 194
4.6 Winter Hourly Urban and Outlying CO2 Concentrations at 2 m above Ground ... 195
4.7 Summer Hourly Urban and Outlying CO2 Concentrations at 0.5 m above
Ground ...... 196
4.8 Winter Hourly Urban and Outlying CO2 Concentrations at 0.5 m above Ground 197
xi Figure Page
5.1 CSTR Chambers in UC Riverside Greenhouse ...... 244
5.2 Growth Rate (mg day-1) and Photosynthetic Yield of S. arabicus Grown in Ambient
and Elevated CO2 and O3 ...... 245
5.3 Growth Rate (mg day-1) and Photosynthetic Yield of P. recurvata Grown in
Ambient and Elevated CO2 and O3 ...... 246
5.4 Observed Percent Relative Change in Growth Rate ...... 247
5.5 Additive and Non-Additive Effects Based on Predicted and Observed Percent
Relative Change in Growth Rate of S. arabicus and P. recurvata ...... 248
xii CHAPTER 1
INTRODUCTION
Cities are interlinked social and ecological systems. The structure and function of the urban environment are inextricably connected with people and their actions (Grimm et al.
2000; Pickett et al. 2001). Urban ecosystems are currently home to more than half the world’s population, though they occupy only a small percentage of Earth’s terrestrial surface (United Nations 2012). Further, by 2030, urban land cover is expected to nearly triple, and by 2050, over 65% of the world’s population will be city dwellers (Seto,
Güneralp, and Hutyra 2012; United Nations 2012). Humans concentrated in cities are the direct and indirect drivers of local and global change, altering ecosystem structure and function, and in turn, ecosystem services across multiple scales (Vitousek et al. 1997).
Thus, understanding the functioning of these complex human-environment systems is of growing environmental, social, and economic concern.
Ecosystems are rarely exposed to a single stressor, particularly in and around cities.
Yet, the ecological impacts in urban and near-urban areas from multiple environmental stressors and their interacting effects are largely unknown. In order to better protect and manage ecosystems in the face of global changes, there is an urgent need to develop a more complete understanding and generalizable theory of cities and their social- ecological feedbacks using comparative approaches within and among cities (Grimm,
Faeth, et al. 2008; Grimm, Foster, et al. 2008; Boone et al. 2012; Bettencourt 2013). This understanding of urban social-ecological functioning must also incorporate the complex effects and interactions of chronic and multiple global change factors resulting from
1 human activities (Zavaleta et al. 2003; Grimm, Faeth, et al. 2008; M. D. Smith, Knapp, and Collins 2009; Hidy and Pennell 2010). Using urban and surrounding native ecosystems as a model system, I address these gaps by examining the complex drivers and direct and indirect ecological outcomes of human activities on both managed landscapes and protected native ecosystems in and around cities.
SCOPE AND STRUCTURE OF THIS DISSERTATION
In Chapter 2, in collaboration with social and natural scientists, I examine the ecological effects of people’s management choices in residential yards and the social- biophysical drivers of those decisions. Residential properties compose nearly half of heterogeneous urban land area and can be hotspots of nutrient inputs and non-point source pollution into urban and surrounding systems. Often highly managed ecosystems where people directly interact with their outdoor environment, residential properties are ideal model systems for examining complex human-environment feedbacks. With an interdisciplinary approach I synthesize the growing body of literature on residential landscapes, including the social drivers of management practices, the ecological outcomes, and the social-ecological feedbacks and tradeoffs. Through this synthesis, I develop a conceptual approach to guide future research and understanding of these complex social-ecological systems. This work was completed in close collaboration with
Kelli Larson (Schools of Sustainability and Geography and Urban Planning, Arizona
State University) and Sharon Hall (School of Life Sciences, Arizona State University).
2 In Chapters 3 – 5, I examine the urban “ecological airshed” and the indirect impacts of human activities via urban-generated air pollutants on protected native ecosystems within and outside of the urban boundary. Atmospheric reactive nitrogen (N), ozone
(O3) and carbon dioxide (CO2) are elevated near human activities. Individually, elevated
N and CO2 act as a resource stimulating primary production, while O3 is a stressor and inhibits production. Urban air quality is expected to have significant impacts on protected lands in urban and surrounding native ecosystems, yet the co-occurring distribution of N, O3, and CO2 in protected lands is unknown. Further, little is known about their combined impacts, and possible non-additive synergistic (amplifying or greater than the sum of the individual effects) or antagonistic (canceling or less than the sum of the individual effects) effects on ecosystem responses at realistic and predicted future concentrations in cities and more remote regions.
For this research, I focus on Phoenix, Arizona and the surrounding Sonoran Desert as a case study and part of the Central Arizona-Phoenix Long-term Ecological Research
(CAP LTER) project. Arid and semi-arid ecosystems cover over a third of the world’s terrestrial land and human population growth and urban expansion are occurring more rapidly in dryland ecosystems than other ecosystem types (MEA 2005; United Nations
2012). The Phoenix metropolitan area, situated in the northern Sonoran Desert, is home to over 4 million people who have significant impacts on ecosystems and air quality.
In addition, this research builds on previous unexpected findings from the Sonoran
Desert that reveal limited responses by herbaceous vegetation to elevated urban N deposition, even in rainy years (Hall et al. 2011). Furthermore, Schmoker/Davis and colleagues (In prep) found that while herbivory accounted for 30% loss of winter
3 herbaceous biomass in dry years, the rates of herbivory are not significantly higher in the urban than outlying locations. Thus, herbivory in cities does not account for the unexpected findings reported by Hall and colleagues (2011). Together, these results suggest that other co-occurring factors related to urbanization and human activity (e.g., elevated CO2, O3, or temperature) may play a significant role in modulating annual plant production. I test this by examining the distribution of ecologically relevant pollutants and their combined impacts on the dominant Sonoran Desert herbaceous species.
In Chapter 3, I specifically investigate the spatial and temporal distribution of N deposition to an arid ecosystem. Anthropogenic activities have doubled the rate of atmospheric reactive N inputs to many ecosystems worldwide with potentially significant ecological implications for biogeochemical cycling and biological diversity in recipient ecosystems (Vitousek et al. 1997; Galloway et al. 2008; Bobbink et al. 2010). Yet, N deposition to dryland ecosystems is not well characterized due to challenges in quantifying dry deposition and inputs from spatially and temporally patchy precipitation
(Fenn et al. 2009). Using multiple sampling approaches, I examine the spatial and temporal patterns and dominant drivers of total wet and dry N deposition in urban and more remote outlying desert locations. Accurately estimating the rate and distribution of
N deposition in native ecosystems is essential for determining where ecosystems exceed the critical load, the level at which negative ecological effects occur.
In Chapter 4, I focus more broadly on the “ecological airshed” created by multiple urban pollutants. These ecologically important atmospheric compounds are often transported into native ecosystems beyond the urban political boundaries in which they are generated and regulated. Some urban air pollutants (e.g. O3 and O3 precursors) are
4 regulated in cities and routinely monitored in dense urban areas for their human health implications. However, the extent of other ecologically significant atmospheric compounds, such as ground-level CO2 and reactive N, within and outside of urban areas is less well known. Identifying the land area affected by the co-distribution of these atmospheric compounds is an important step for setting effective management and conservation strategies to protect native ecosystems and the ecosystem services they provide. Specifically, I monitor local and regional patterns of CO2, O3, and reactive gaseous N concentrations and deposition in native protected desert ecosystems in the urban and outlying regions of the Phoenix metropolitan area. In addition, I examine the small-scale variability of N and O3 concentrations along a transect from the exterior to interior of a large protected desert area within the city. I expect areas in the interior to more closely resemble outlying native ecosystems with decreasing pollutant concentrations farther from the source (i.e. exterior edge). This study is the first to identify the distinct spatial pattern of co-occurring, ecologically important urban pollutants within protected lands.
In Chapter 5, I examine the combined ecological effects of co-occurring CO2, N, and
O3. Though the individual effects of CO2, N, and O3 are well understood, it is uncertain if their combined impacts will cause non-additive (i.e. synergistic or antagonistic) ecosystem responses, which are more difficult to predict and model. In a multi-factor dose-response greenhouse experiment, I investigate the net growth and physiological responses of dominant native and non-native desert species found ubiquitously throughout the native desert near Phoenix, Arizona. Specifically, I test the combined impact of CO2, N, and O3 at levels that are reflective of current air quality conditions in
5 Phoenix, as well as predicted future air quality under current rates of global change
(IPCC 2014). With the full factorial experimental design, I compare the potential for synergistic or antagonistic responses in multi-factor treatments. Ultimately, this research will provide empirical evidence relevant for future management decisions for preserving native ecosystems within the “airshed” affected by co-occurring pollutants.
Finally, in Chapter 6, I briefly synthesize the overarching findings and main contributions from this research, as well as suggest some next steps and implications for future management.
6 REFERENCES
Bettencourt, Luís M. A. 2013. “The Origins of Scaling in Cities.” Science 340 (6139): 1438–41. doi:10.1126/science.1235823.
Bobbink, Roland, K. Hicks, J. Galloway, T. Spranger, R. Alkemade, M. Ashmore, M. Bustamante, S. Cinderby, E. Davidson, and F. Dentener. 2010. “Global Assessment of Nitrogen Deposition Effects on Terrestrial Plant Diversity: A Synthesis.” Ecological Applications 20 (1): 30–59.
Boone, Christopher G., Elizabeth Cook, Sharon J. Hall, Marcia L. Nation, Nancy B. Grimm, Carol B. Raish, Deborah M. Finch, and Abigail M. York. 2012. “A Comparative Gradient Approach as a Tool for Understanding and Managing.” Urban Ecosystems 15 (4): 795–807. doi:10.1007/s11252-012-0240-9.
Galloway, James N., Alan R. Townsend, Jan Willem Erisman, Mateete Bekunda, Zucong Cai, John R. Freney, Luiz A. Martinelli, Sybil P. Seitzinger, and Mark A. Sutton. 2008. “Transformation of the Nitrogen Cycle: Recent Trends, Questions, and Potential Solutions.” Science 320 (5878): 889–92.
Grimm, Nancy B., Stanley H. Faeth, Nancy E. Golubiewski, Charles L. Redman, Jianguo Wu, Xuemei Bai, and John M. Briggs. 2008. “Global Change and the Ecology of Cities.” Science 319 (5864): 756–60. doi:10.1126/science.1150195.
Grimm, Nancy B., David Foster, Peter Groffman, J. Morgan Grove, Charles S. Hopkinson, Knute J. Nadelhoffer, Diane E. Pataki, and Debra PC Peters. 2008. “The Changing Landscape: Ecosystem Responses to Urbanization and Pollution across Climatic and Societal Gradients.” Frontiers in Ecology and the Environment 6 (5): 264–72.
Grimm, Nancy B., J. M. Grove, S. T. A. Pickett, and C. L. Redman. 2000. “Integrated Approaches to Long-Term Studies of Urban Ecological Systems.” Bioscience 50 (7): 571–84. doi:10.1641/0006-3568(2000)050[0571:IATLTO]2.0.CO;2.
Hall, Sharon J., Ryan A. Sponseller, Nancy B. Grimm, David Huber, Jason P. Kaye, Christopher Clark, and Scott L. Collins. 2011. “Ecosystem Response to Nutrient Enrichment across an Urban Airshed in the Sonoran Desert.” Ecological Applications 21 (3): 640–60. doi:10.1890/10-0758.1.
Hidy, George M., and William T. Pennell. 2010. “Multipollutant Air Quality Management.” Journal of the Air & Waste Management Association 60 (6): 645– 74. doi:10.3155/1047-3289.60.6.645.
IPCC. 2014. “Intergovernmental Panel on Climate Change: Fifth Assessment Report.”
7 http://www.ipcc.ch/.
MEA. 2005. “Millennium Ecosystem Assessment: Ecosystems and Human Well-Being.” Island Press. http://www.unep.org/maweb.
Pickett, S. T. A., M. L. Cadenasso, J. M. Grove, C. H. Nilon, R. V. Pouyat, W. C. Zipperer, and R. Costanza. 2001. “Urban Ecological Systems: Linking Terrestrial Ecological, Physical, and Socioeconomic Components of Metropolitan Areas.” Annual Review of Ecology and Systematics 32: 127–57. doi:10.1146/annurev.ecolsys.32.081501.114012.
Seto, Karen C., Burak Güneralp, and Lucy R. Hutyra. 2012. “Global Forecasts of Urban Expansion to 2030 and Direct Impacts on Biodiversity and Carbon Pools.” Proceedings of the National Academy of Sciences 109 (40): 16083–88. doi:10.1073/pnas.1211658109.
Smith, Melinda D., Alan K. Knapp, and Scott L. Collins. 2009. “A Framework for Assessing Ecosystem Dynamics in Response to Chronic Resource Alterations Induced by Global Change.” Ecology 90 (12): 3279–89. doi:10.1890/08-1815.1.
United Nations. 2012. “World Urbanization Prospects, The 2011 Revision”. United Nations publication.
Vitousek, Peter M., John D. Aber, Robert W. Howarth, Gene E. Likens, Pamela A. Matson, David W. Schindler, William H. Schlesinger, and David G. Tilman. 1997. “HUMAN ALTERATION OF THE GLOBAL NITROGEN CYCLE: SOURCES AND CONSEQUENCES.” Ecological Applications 7 (3): 737–50. doi:10.1890/1051-0761(1997)007[0737:HAOTGN]2.0.CO;2.
Zavaleta, Erika S., M. Rebecca Shaw, Nona R. Chiariello, Brian D. Thomas, Elsa E. Cleland, Christopher B. Field, and Harold A. Mooney. 2003. “GRASSLAND RESPONSES TO THREE YEARS OF ELEVATED TEMPERATURE, CO2, PRECIPITATION, AND N DEPOSITION.” Ecological Monographs 73 (4): 585–604. doi:10.1890/02-4053.
8 CHAPTER 2
RESIDENTIAL LANDSCAPES AS SOCIAL-ECOLOGICAL SYSTEMS: SYNTHESIS
OF MULTI-SCALAR INTERACTIONS BETWEEN PEOPLE AND THEIR HOME
ENVIRONMENT
AUTHORS: Elizabeth M. Cook, Sharon J. Hall, Kelli L. Larson
PUBLISHED: 2012, Urban Ecosystems, 15:19-52
ABSTRACT
Residential landscapes are a common setting of human-environment interactions. These ubiquitous ecosystems provide social and ecological services, and yard maintenance leads to intended and unintended ecological outcomes. The ecological characteristics of residential landscapes and the human drivers of landscape management have been the focus of disciplinary studies, often at a single scale of analysis. However, an interdisciplinary examination of residential landscapes is needed to understand the feedbacks and tradeoffs of these complex adaptive social-ecological systems as a whole.
Our aim is to synthesize the diversity of perspectives, scales of analysis, and findings from the literature in order to 1) contribute to a holistic, interdisciplinary understanding of residential landscapes and 2) identify research needs while providing a robust conceptual approach for future studies. We synthesize 256 studies from the literature and develop an interdisciplinary, multi-scalar framework on residential landscape dynamics.
9 From our synthesis, we find that complex human drivers, including attitudinal, structural, and institutional factors at multiple scales, influence management practices, which in turn determine biophysical characteristics of residential landscapes. However, gaps exist in our interdisciplinary understanding of residential landscapes within four key but understudied areas: 1) the link between social drivers and ecological outcomes of management decisions, 2) the ecosystem services provided by these landscapes to residents, 3) the interactions of social drivers and ecological characteristics across scales, and 4) generalizations of patterns and processes across cities. Our holistic perspective will help to guide future interdisciplinary collaborations to integrate theories and research methods across geographic locations and spatial scales.
10 INTRODUCTION
Residential landscapes are a primary setting of everyday interactions between humans and the environment (Bhatti and Church 2001). Approximately 75% of people in developed regions live in urbanized areas, and an estimated 41% of urban land area is used for homes and their surroundings (UN 2010; Nowak et al. 1996). Within the outdoor area surrounding homes (hereafter, “landscapes” or “yards”), turfgrass yards are highly managed ecosystems that rival corn as the most extensive irrigated crop in the US
(Milesi et al. 2005). Along with gardens and more “natural” appearing lawn alternatives, these landscapes embody the idealized preferences and socially constrained practices of residents (Jenkins 1994; Robbins 2007). Similar to well studied agricultural systems
(e.g., Matson et al. 1998; Tilman et al. 2002), residential landscapes provide important amenities while contributing to both intended and unintended environmental consequences (Martin 2008; Larson et al. 2009a). Residential landscapes are complex adaptive systems (sensu Holland 1995), in which multiple social and biophysical processes and feedbacks occur at a range of scales, from parcels and neighborhoods to watersheds and larger-scale regions. Thus, understanding the tradeoffs and feedbacks associated with these coupled human-natural systems necessitates interdisciplinary research that considers a variety of spatial scales.
Both social and natural science research have long explored human-environment interactions using disciplinary approaches. Previous residential landscape studies have been conducted primarily from social (e.g., Askew and McGuirk 2004) or ecological
(e.g., Sperling and Lortie 2010) perspectives, typically focusing on a single scale of
11 analysis. At the household-scale, for instance, social scientists have studied residents’ preferences for certain yard features (e.g., Larson et al. 2009a), while ecologists have explored biodiversity (e.g., Smith et al. 2006c) often with social surveys and observational field studies, respectively. Broader-scale studies have examined residential land-use/land-cover patterns, commonly with remote sensing imagery and geospatial technology in single regions (e.g., Grove et al. 2006a). The nascent field of urban ecology has begun to examine coupled human-environment interactions using interdisciplinary approaches (Pickett et al. 1997; Collins et al. 2000; Grimm et al. 2000), with limited but increasing attention to residential landscapes (Byrne and Grewal 2008).
However, disparate disciplinary perspectives, analytical methods, and different scales of analysis render generalizations difficult and limit an integrated understanding of residential landscape dynamics.
Here, our aim is to synthesize disciplinary perspectives and scales of analysis from the literature in order to 1) contribute to a holistic, integrated understanding of the causes, consequences, and feedbacks related to residential landscapes as complex social- ecological systems, and 2) identify compelling research directions while providing a robust conceptual approach to guide future studies. We, thus, critically review the growing body of residential landscapes literature to synthesize the social-ecological interconnections across scales within this system. From our review, we examine temporal trends of research regarding residential landscapes. We also develop an interdisciplinary framework that encompasses multiple scales and disciplinary perspectives to guide future interdisciplinary research on the human drivers of landscaping practices and the ecological outcomes of those landscaping decisions. Our
12 interdisciplinary approach advances a holistic understanding of feedbacks and tradeoffs associated with residential landscapes as a model social-ecological system, and it provides a guide for future integrated research and theories. In the following sections, we synthesize findings from the literature across the system components at the household
(parcel), neighborhood, and broader-scales, including: the ecological properties, function and ecosystem services of residential landscapes; the land management decisions that create and maintain these ecosystems; the social drivers—attitudinal, structural, and institutional—of those practices; and long-lasting legacies from past decisions that influence yard structure, management, and ecosystem services. Finally, we conclude by highlighting knowledge gaps and suggesting directions for future research on residential landscapes as complex human-environment systems.
METHODS
To review publications related to residential landscapes, we searched titles, abstracts and author keywords within Web of Science, EbscoHost, and other relevant journals not included in these databases (e.g., Urban Ecosystems) using the following a priori key terms (alone and in combination): resident*, hous*, yard, garden, landscap*, lawn, turf*.
Asterisks indicate partial search words for which multiple word derivations may be relevant (e.g., resident and residential). Terms were searched in several combinations using “AND” and “OR” statements, primarily by combining terms to represent our interest in residential households and yards (e.g., resident* OR hous* AND yard OR garden OR landscap* OR lawn). Keyword searches were repeated in multiple
13 combinations until no new relevant publications were found. Along with exhaustive searches, we identified additional publications from references of articles obtained in the database searches. Articles that did not specifically address or examine data regarding privately managed residential properties and their residents were excluded, as they were beyond the scope of our research. This focus narrows our review to mostly outdoor spaces of developed urban and suburban regions where single-family residences dominate. It thus excludes some residential settings (e.g., apartment complexes) that do not have private outdoor landscapes. In total, 256 studies were critically reviewed.
For each article, we first identified the dominant research question(s), then grouped publications by related topics and reviewed them for themes, results, and methodological approaches. Because our intent is to advance an interdisciplinary understanding of residential landscapes, we focused on the commonalities, differences, strengths, and weaknesses from the literature collectively, while identifying areas for future research.
In this paper, we reviewed “multidisciplinary” studies through which several different fields explore questions or system dynamics from single disciplinary perspectives. We use the term “interdisciplinary” specifically to encompass research that integrates more than one disciplinary approach in addressing questions collectively from both social and ecological perspectives. In addition to examining the change in multidisciplinary and interdisciplinary research on residential landscapes over time, we classified articles into three overarching categories, natural science, social science, or both (interdisciplinary), based on the article’s research objective and methods. For example, research examining fertilizer use is classified as “social science” if residents were surveyed or interviewed about their landscaping practices or personal attributes (e.g. demographics); as “natural
14 science” if ecological properties were examined as a result of fertilization; or
“interdisciplinary” if a combination of methods or approaches was used.
In the preliminary stages of our research, we initially conceived of a framework encompassing relevant disciplinary and interdisciplinary perspectives. The initial conception drew on social and ecological approaches, ranging from human-ecological theories of behavior (e.g., the value-belief-norm model; Stern 2000) to broad interdisciplinary frameworks in urban ecology (e.g., highlighting key social and ecological patterns and processes, such as demographics and institutions, as well as ecosystem functions and services; Grimm et al. 2000; Pickett et al 2001; Redman et al.
2004). We then further developed and refined the framework’s system components
(boxes and arrows) based on our literature review specific to residential landscapes. The framework synthesizes the major relationships emerging from theoretical and empirical evidence in the published literature on the social-ecology of yard management. The framework focuses on the bi-directional interactions between the drivers of residents’ practices, the resulting ecological properties and processes, ecosystem services, legacy effects, and the feedbacks that reinforce or constrain land management. While such linkages could be addressed for other social-ecological systems or contexts, such as the management of public parks, we maintain a focus on privately owned residential landscapes as a prominent everyday setting for human-environment interactions in urban ecosystems.
15 FINDINGS
Ecology of residential landscapes
Cities are heterogeneous ecosystems with complex ecological properties and processes (Band et al. 2005). Humans directly and indirectly impact urban—and residential—ecological characteristics through management of built and biotic infrastructure as well as alteration of climate conditions and food webs, among other ecosystem elements. Most studies on residential landscapes have been conducted using methods from natural science disciplines (68% of all 256 studies; Fig. 1) with a focus on ecological properties of yards at the household-scale (56% of 174 ecology studies from natural science and interdisciplinary papers). Fewer studies explore ecological functioning and services (41% of 174 ecology studies), or ecological properties at larger scales (27% of 174 ecology studies).
Ecological properties of residential landscapes: The biotic and abiotic physical characteristics of yards include various groundcovers, species composition and abundance, soil properties and microclimates (Fig. 2a, Ecological Properties).
Residential landscapes cover approximately a quarter of the land within cities, and nearly half of residential land is vegetated (Gaston et al. 2005; Mathieu et al. 2007). While biophysical properties are heterogeneous within and between individual residential properties, they follow predictable patterns based primarily on yard size, housing density and age, homeowner socioeconomic and lifestyle factors, management practices, vegetation composition, and legacies of former land use (Table 1). At broader scales
16 across cities, residential biophysical patterns are commonly examined in relation to regional climate and aggregated household and social data.
Ecological properties at the household-scale: Green lawns with shade trees are often perceived as a homogeneous manifestation of “the American Dream,” as they are ubiquitous across United States (US) cities in diverse biomes (Jenkins 1994; Bormann et al. 2001). Yet at the parcel-scale, residential landscape structure varies considerably within and between neighborhoods (Martin et al. 2003; Gaston et al. 2005; Crow et al.
2006; Kirkpatrick et al. 2007; Luck et al. 2009; yet see Loram et al. 2008b).
Groundcovers vary between grass, bare soil, rocks, and impervious surfaces (e.g. pools, patios, parking areas), with numerous grass, herb, shrub, and tree species (Henderson et al. 1998; Zmyslony and Gagnon 1998; Martin et al. 2003; Thompson et al. 2004; Daniels and Kirkpatrick 2006a; Smith et al. 2006c; Loram et al. 2008a; Luck et al. 2009). To understand this seemingly random suite of species and structures, investigators have classified yards into distinct morphologies. In an arid region, Martin and colleagues
(2003) identified three yard types based on vegetation and water-use intensity: “mesic” yards with turfgrass and shade trees, “xeric” yards with gravel and drought-adapted plants, and “oasis” yards that contain both mesic and xeric features. In a temperate climate, Daniels & Kirkpatrick (2006a) identified distinct categories based on vegetation architecture, including height and percent cover. A number of studies show that parcel- scale landscape variability is less random than perceived. For example, the area of yard structures such as patios, cultivated borders, and vegetation canopy is positively correlated to yard size (Table 1; Richards et al. 1984; Smith et al. 2005; Daniels and
Kirkpatrick 2006a; Smith et al. 2006c; Kirkpatrick et al. 2007; Loram et al. 2008b) and
17 socioeconomic characteristics of homeowners (Martin et al. 2004; Kirkpatrick et al.
2007; Luck et al. 2009), and it is negatively correlated with housing density (Smith et al.
2005; Marco et al. 2008).
Plant community composition also varies among households, but patterns emerge across studies. Non-native species that originate from many parts of the world dominate the flora of residential landscapes, making up 67-88% of all woody species (Livingston et al. 2003; Thompson et al. 2003; Smith et al. 2006c; Acar et al. 2007; Loram et al. 2008a;
Marco et al. 2008). However, indigenous lawn species are far greater in European cities and tropical regions than elsewhere (Stewart et al. 2009; Akinnifesi et al. 2010). Most individual exotics are reported at very low frequencies, suggesting an overall high diversity and turnover of non-native species (Thompson et al. 2003; Acar et al. 2007;
Loram et al. 2008a; Marco et al. 2008). Although species richness and diversity are often positively related to yard size (Thompson et al. 2004; Smith et al. 2006c), this relationship may vary by region, as it is not consistent across studies (Table 1;
Albuquerque et al. 2005; Stewart et al. 2009). Finally, in some instances species composition in yards is related to housing age (Smith et al. 2005; Acar et al. 2007). For example, in Turkey, younger housing had a greater variety of ornamental species and older residential areas had traditional, functional gardens (e.g., fruiting trees; Acar et al.
2007).
Most residential vegetation studies focus on front yards because they are readily surveyed through field observations; however, a few highlight differences in front and backyard biotic structure (Dorney et al. 1984; Richards et al. 1984; Daniels and
Kirkpatrick 2006a; Loram et al. 2007). Although the paucity of data prevents
18 generalization, front yards are often more highly manicured than backyards, likely due to their visibility to the public (Richards et al. 1984; Daniels and Kirkpatrick 2006a; Larsen and Harlan 2006; Larson et al. 2009a). In contrast, backyards are more likely to contain flower and vegetable gardens and a greater extent of grass than front yards (Richards et al. 1984; Daniels and Kirkpatrick 2006a). However, newer suburban developments tend to have larger houses and smaller backyards than in older communities, which may affect how the backyard is utilized and managed (Hall 2010).
Patterns in vegetation and management practices at the household-scale have diverse, often taxon-specific, consequences for the abundance and diversity of invertebrate species that utilize yards as habitat or resources (Livingston et al. 2003; Raupp et al.
2010). For instance, yard habitat structure influences the abundance of ground-dwelling organisms (e.g., Byrne 2007). Some ground arthropods, such as spiders and harvestmen, thrive in mesic (grass) yards compared to other yard types, likely due to the productivity and resource availability associated with irrigation (Shochat et al. 2004; Cook and Faeth
2006). However, lawns with varying management (e.g., chemical versus no chemical inputs) exhibit little difference in arthropod (e.g., Collembolas) and nematode abundance
(Byrne and Bruns 2004; Cheng et al. 2008). Abundance of ground invertebrates, such as earthworms, is inversely related to soil bulk density (Smetak et al. 2007; Byrne et al.
2008) while some winged and non-winged invertebrate abundance is positively correlated with aboveground characteristics such as vegetation richness and plant structure (Table 1;
Smith et al. 2006b; Sperling and Lortie 2010).
Similar to some natural ecosystems (Waide et al. 1999; Mittelbach et al. 2001), residential faunal diversity can be negatively related to plant productivity. Invertebrate
19 species richness is often lower in lawns than in less productive xeric yards (McIntyre and
Hostetler 2001; Shochat et al. 2004; yet see Cook and Faeth 2006), although species richness increases near the lawn’s edge where species mix from adjacent habitats (e.g., shrub borders; Smith et al. 2006a). Additionally, plant species composition within yards, for example native versus non-native species, is especially important to butterfly, bee, and bird populations (Table 1; Germaine et al. 1998; Marzluff and Ewing 2001; Yahner
2001; Chace and Walsh 2006; Daniels and Kirkpatrick 2006b; French et al. 2005;
Fetridge et al. 2008; Aurora et al. 2009; Burghardt et al. 2009).
Urban biotic and abiotic characteristics, as well as human actions, affect soil physical and chemical properties, leading to a divergence from native soils (e.g., Golubiewski
2006; Pouyat et al. 2007). Pickett and Cadenasso (2009) characterized urban soils based on the five factors of soil formation (Jenny 1941)—soil parent material, the effect of time, species composition, topography, and microclimate—in addition to human impacts.
Parent material of urban and residential areas is generally distinguished from native soils by disturbances from urban development and construction, resulting in a mixed substrate and compaction (Effland and Pouyat 1997; Lehmann and Stahr 2007). Over time, older residential soils tend to have lower bulk density and contain larger total pools of organic carbon (C) and nitrogen (N) than newer residential properties, regardless of location across the US (Table 1; Law et al. 2004; Scharenbroch 2005; Golubiewski 2006; Smetak et al. 2007; Pouyat et al. 2009).
Species composition, current and past land uses, and human management also impact residential soil characteristics and related microclimates. Soil organic C storage is generally higher in residential yards compared to surrounding native forests (Pouyat et al.
20 2009), grasslands (Kaye et al. 2005; Golubiewski 2006), and deserts (Jenerette et al.
2006; Kaye et al. 2008), and it is positively correlated with fertilization, irrigation, and returning grass clippings (Pouyat et al. 2002; Qian et al. 2003; Pouyat et al. 2009). Plant available phosphorus (P) is greater in lawns than native prairies, but lower than in other cultivated lands such as farms (Bennett et al. 2005). In addition, residential soils are affected by legacies of former land cover or land use decisions. Due to antecedent soil fertility, soils from lawns on former agricultural fields contain larger pools of soil organic matter, C, N, and bioavailable P than soils from residential properties that were previously desert (Hope et al. 2005; Lewis et al. 2006; Davies and Hall 2010). Finally, plant communities, landscape characteristics, and management practices such as irrigation combine to create cooler microclimates through evapotranspiration (Bonan
2000; Coutts et al. 2007; Jenerette et al. 2007; Tratalos et al. 2007b; Baris et al. 2009;
Shashua-Bar et al. 2009; Peters and McFadden 2010), which in turn regulates soil moisture (Trudgill et al. 2010).
Ecological properties at neighborhood and broader-scales: When aggregated, heterogeneous ecological properties within parcels lead to broad patterns at regional and larger scales. Houses clustered in close proximity within a neighborhood often share related yard characteristics (Zmyslony and Gagnon 1998, 2000; but see Kirkpatrick et al.
2009). Vegetation cover is inversely associated with housing and population density
(Iverson and Cook 2000; Marco et al. 2008; Luck et al. 2009; Boone et al. 2010).
Additionally, both floral diversity and cover are inversely related to neighborhood age
(Martin et al. 2004; Hope et al. 2006; yet see Grove et al. 2006b who report a quadratic relationship). This finding is contrary to the fundamental ecological theory of island
21 biogeography, which predicts a positive relationship between age of site and diversity
(MacArthur and Wilson 2001). Vegetation diversity and cover across neighborhoods is positively related to socioeconomic advantage, education, and lifestyle factors that reflect group identity and social status (Table 1; Iverson and Cook 2000; Hope et al. 2003;
Martin et al. 2004; Grove et al. 2006b; Mennis 2006; Tratalos et al. 2007b; Luck et al.
2009; Boone et al. 2010). Similarly, invasive plant richness is positively related to income, low-density housing, and area of urban-wildland interface at the county scale
(Gavier-Pizarro et al. 2010). Despite general patterns, the geographic context and historic processes of urban growth are important factors in predicting current vegetation trends (Luck et al. 2009; Boone et al. 2010). In Detroit, Michigan, for example, vegetation cover was positively associated with exurban growth and economic development, but also neighborhood abandonment in low-income, inner-city neighborhoods (Ryznar and Wagner 2001). At the city scale, US residential tree cover is dictated by climate as it varies predictably by biome, from 31% in forested cities, 19% in grassland cities to 17% in desert cities (Nowak et al. 1996).
The distribution and composition of fauna across residential neighborhoods, particularly species with large ranges, are negatively affected by landscape patchiness and fragmentation. As predicted by the theory of island biogeography, bird, butterfly, and small mammal abundance and diversity increase with habitat area (Chamberlain et al.
2004; Daniels and Kirkpatrick 2006b; Baker and Harris 2007; Evans et al. 2009), neighborhood age (Edgar and Kershaw 1994; Yahner 2001; yet see Loss et al. 2009), and proximity to natural habitats, source populations, and corridors (Germaine et al. 1998;
Daniels and Kirkpatrick 2006b; Baker and Harris 2007; Loss et al. 2009). In addition,
22 residential landscapes in conjunction with local open, green spaces may form a large urban matrix of habitats, corridors and resources that can provide beneficial services to urban species (James et al. 2009; Goddard et al. 2010). At larger city-scales, faunal species abundance is often greater in urban areas, while diversity is lower or different in composition compared to surrounding native ecosystems (Marzluff and Ewing 2001;
Shochat et al. 2004; Chace and Walsh 2006; Cook and Faeth 2006; McKinney 2006;
Catterall et al. 2010). While some common trends exist, patterns often vary by species, region, or method and scale of analysis (Smith et al. 2006b; Smith et al. 2006d). For instance, bird, butterfly and small mammal abundance and richness vary between positive, negative and non-linear relationships with housing density (Germaine et al.
1998; Germaine et al. 2001; Yahner 2001; Baker and Harris 2007; Tratalos et al. 2007a;
Evans et al. 2009; Hourigan et al. 2010).
Ecological functioning and ecosystem services of residential landscapes: Biophysical properties regulate material and energy flow between humans, non-human biota, soils, water, the atmosphere, and other ecosystems (Fig. 2b, Ecological Function). Ecological functions include cycling of water and elements, as well as trophic dynamics. Together, ecosystem properties and functions provide many services to people (MEA 2005; Fig. 2c,
Ecosystem Services). Like their native counterparts, managed landscapes regulate ecological properties and processes such as microclimate and pollination (regulating services), provide support for other services through primary production and nutrient cycling (supporting services), and produce goods such as food resources (provisioning services; Beard and Green 1994). In addition, residential yards offer key, social amenities such as a sense of place for people and communities (cultural services).
23 Finally, human actions and natural causes (e.g., storms) are sometimes responsible for negative ecosystem services, or “disservices” such as air or water pollution, that result in social and ecological costs (Tratalos et al. 2007b; Grimm et al. 2008).
To date, most studies on ecological functioning in yards have been conducted at the parcel-scale (86% of 72 studies focused on ecological processes), with the exception of some research on surface temperatures (Jenerette et al. 2007) and watershed-scale nutrient budgets (Groffman et al. 2004). While residential outdoor spaces can provide numerous services to residents, few studies explicitly examine the ecosystem services and disservices of residential landscapes. Collectively, studies show that ecosystem processes in residential landscapes often differ in rate, magnitude or variability from those in surrounding native biomes due to human impacts on yard composition and water and nutrient availability through management practices (Table 2).
Irrigation, species composition, and plant physiological processes affect water fluxes among residential yards and other ecosystem components, such as the atmosphere.
Similar to native ecosystems, the areal extent of residential vegetated ground and canopy cover is positively related to evapotranspiration—the loss of water from soil and vegetation to the surrounding atmosphere (Bonan 2000; Martin et al. 2007; Shashua-Bar et al. 2009). Evapotranspiration creates a positive service, cooling air around homes and potentially reducing energy use for air conditioning (Huang et al. 1987; Bonan 2000;
Coutts et al. 2007; Georgi and Dimitriou 2010). One study estimated air conditioning costs can be reduced by approximately 2% per residential tree (Simpson and McPherson
1998), although tree species vary in their effects on human thermal comfort (Georgi and
Dimitriou 2010). Likewise, irrigated mesic landscapes mitigate surface temperatures
24 better than impervious surfaces or less-irrigated yards (Bonan 2000; Martin et al. 2007).
At the neighborhood scale, high pavement cover and housing density are positively related to warmer land surface termperatures, while increased vegetation cover cools temperatures and may reduce human heat stress and other urban heat island effects
(Stabler et al. 2005; Harlan et al. 2006; Jenerette et al. 2007; Buyantuyev and Wu 2010;
Su et al. 2010).
Potential tradeoffs between ecosystem services exist, for example, between water and energy use for varying types of residential landscapes. While highly productive yards might mitigate energy use and heat stress, they require more water and, in turn, intensive pruning that incurs the cost of labor and time (Martin 2008). Yards with drought-adapted vegetation require less water than lawns, but irrigation of xeric yards is often not adjusted for seasonal changes in evapotranspiration and results in over-watering (Martin 2001;
Sovocool et al. 2006; St Hilaire et al. 2008). Moreover, high irrigation rates and pruning can cause vegetation to use water less efficiently, as well as decrease the potential to store carbon (Stabler 2008; Martin 2008). Recognizing the tradeoffs between lawns and xeric yards, some residents intentionally choose “oasis”-type yards to achieve “best of both worlds” benefits in desert cities, such as Phoenix, Arizona (Larson et al. 2009a).
Like all ecosystems, residential landscapes support trophic exchange between non- human organisms through predation and herbivory. Species’ mobility and resource demands, along with the variety of habitats and human management practices make it difficult to identify trends across yards and for specific species (Smith et al. 2006b; Smith et al. 2006d). However, managed landscapes generally contain more abundant and less seasonally variable resources than many native ecosystems (Shochat et al. 2006),
25 particularly with supplementary nest boxes and feeders (Gaston et al. 2005; Fuller et al.
2008; Davies et al. 2009). Thus, competitive interactions and top-down versus bottom-up trophic controls are altered, which in turn shifts faunal species composition and predation rates (Faeth et al. 2005; Cook and Faeth 2006; Shochat et al. 2006). For instance, synanthropic species (species associated with humans) may exploit urban resources, thereby out-competing non-synanthropic species and preventing them from adapting to the urban environment (Shochat et al. 2010). An increase in urban exploiter species, who likely thrive in highly managed residential yards, may shift community assemblages toward biotic homogenization in cities (McKinney 2006). Additionally, altered urban floral and faunal communities may impact ecosystem services, such as biodiversity, pollination, seed dispersal, and beneficial pest regulation (Andersson et al. 2007; Davies et al. 2009; Doody et al. 2010).
Yard properties and management practices also influence C cycling, a primary life- supporting service. Methane (CH4) is a carbon-based greenhouse gas and is generally consumed by microbial processes in well-drained soils. Rates of CH4 uptake—or consumption—decline with N enrichment associated with fertilization and atmospheric deposition (Groffman and Pouyat 2009). Fertilized lawn soils consume CH4 at lower rates than native soils or can even become a CH4 source—an ecosystem disamenity
(Kaye et al. 2004; Groffman and Pouyat 2009). While lawn CH4 consumption rates are similarly low across studies, they vary little among different irrigation and fertilization treatments in urban lawns (Livesley et al. 2010).
Carbon dioxide (CO2), another greenhouse gas, is a product of metabolic respiration and a primary requirement for photosynthesis. Through microbial and root respiration,
26 soils within some residential groundcover types emit more CO2 than others, likely due to optimal soil temperature, moisture, and organic C available for microbial activity in managed landscapes (Green and Oleksyszyn 2002; Koerner and Klopatek 2002; Byrne et al. 2008). Overall, lawn soil CO2 emissions are high relative to eastern temperate forests, agricultural fields, and native grasslands (Kaye et al. 2005; Groffman et al. 2009).
Through photosynthesis, CO2 is transformed into organic matter that supports plant growth and C storage, particularly in highly irrigated yards and older landscapes with low disturbance and high vegetation productivity (e.g., Golubiewski 2006; Stabler 2008;
Townsend-Small and Czimczik 2010).
Like carbon, N is an essential element for life and is affected by ecological properties and management of yards. At the parcel-scale in arid regions, fertilized lawn soils emit more nitrous oxide (N2O), another greenhouse gas, than soils of native grasslands, deserts, agricultural land, or xeric, rock-covered yards (Kaye et al. 2004; Hall et al. 2008;
Hall et al. 2009). However in temperate climates, rates of denitrification, a microbial process that produces N2O in soils, does not differ between lawn and other organic-rich native forest soils or unmanaged old fields, which may be a result of similar soil moisture
(Byrne 2006; Groffman et al. 2009). N2O emissions from lawn soils are explained by a direct relationship to soil moisture (Bijoor et al. 2008; Hall et al. 2008; Groffman et al.
2009; Livesley et al. 2010), fertilizer application (Bijoor et al. 2008; Livesley et al. 2010;
Townsend-Small and Czimczik 2010), and soil temperature (Bijoor et al. 2008; Hall et al.
2008). Due to the areal extent of fertilized grass lawns within cities and the urban heat island, parcel-scale management practices may contribute substantially to regional biosphere-atmosphere gas dynamics (Bijoor et al. 2008). For example, regular
27 fertilization and irrigation reduces seasonal variation in soil gas fluxes, such as N2O and nitric oxide, a reactive gas that contributes to regional smog (Hall et al. 2008).
- Excess soil nitrate (NO3 ) is a common concern near urban and agricultural areas where leaching and runoff from fertilizer can contaminate groundwater or downstream ecosystems, creating an ecosystem disservice (Petrovic 1990; Groffman et al. 2004).
- Lawn soils tend to leach more NO3 than native forest soils, but less than agricultural soils
- (Groffman et al. 2009). Lawn NO3 leaching is dependent on water and fertilizer applications, soil texture (Petrovic 1990), and turfgrass species (Liu et al. 1997; Erickson et al. 2001). Nitrate leaching rates are particularly high in new yards where management may be more intense (Law et al. 2004; Oki et al. 2007), whereas older lawns with lower runoff rates are better at retaining excess nutrients and mitigating water pollution
(Groffman et al. 2004; Pickett et al. 2008; Raciti et al. 2008). Similarly, grass yards can
- be relatively efficient at retaining NO3 when they have high soil organic matter and actively growing microbial communities characteristic of older, undisturbed soils (Qian et al. 2003; Groffman et al. 2004; Raciti et al. 2008; Groffman et al. 2009).
Arguably, the most important services provided by residential landscapes are cultural, as they promote human mental and physical well-being and are places for recreation and gathering with family and friends (Beard and Green 1994; Matsuoka and Kaplan 2008;
Abraham et al. 2010). Indeed, residents report a sense of comfort in their yards (Crow et al. 2006), as well as greater neighborhood satisfaction and social interactions when surrounded by more open space and trees (Kweon et al. 2010; Uslu and Gokce 2010).
Outdoor space is often considered a functional extension of the home, designed and managed to meet aesthetic and recreational preferences (Jenkins 1994; Bhatti and Church
28 2001; Martin et al. 2003; Larsen and Harlan 2006; Larson et al. 2009a). In many areas of the world, yards provide subsistence “homegardens,” with not only ornamental species, but also supplementary food and medicinal flora (Albuquerque et al. 2005; Eichemberg et al. 2009; Huai and Hamilton 2009; Kabir and Webb 2009). Yards also benefit human health through connections to the outdoors (Bhatti and Church 2001; Fuller et al. 2007;
Tzoulas et al. 2007), and offer a “sense of place,” reminding residents of native ecosystems, their geographic place of origin, or particular settings such as “home” or
“nature” (Larson et al. 2009a). Finally, landscaping is frequently perceived as an indicator of property investment and value, which may lead to community cohesiveness
(Grove et al. 2006b; Robbins 2007).
Management decisions
Residents alter ecological properties and functions of their landscapes by installing and removing vegetation and impervious cover, using various irrigation technologies, mowing, pruning, and determining the timing, amount, and frequency of water and chemical inputs (Fig. 2d, Management Decisions). Thus far, residential landscape management research has mostly been conducted at the household-scale (83% of 39 management studies) through social surveys on landscaping practices and related social drivers. Our literature synthesis shows that management practices vary by the primary caretaker (e.g., resident versus professional service), knowledge of different management strategies, yard cover, irrigation technology (Table 3), and residents’ personal ideals and attributes, such as aesthetic preferences and income (Table 4).
29 Yard management varies between do-it-yourself approaches and professional services with predictable differences in yard “quality.” In surveys from Ohio, North Carolina and
Oregon, approximately 16–43% of households use professional landscaping services
(Robbins et al. 2001; Osmond and Hardy 2004; Nielson and Smith 2005), while 70% of
US households cared for at least one aspect of their own yard (NGA 2007).
Professionally maintained yards often rank higher in aesthetic lawn quality, as measured through grass color (greenness) and a lack of weeds (monoculture), than do-it-yourself approaches (Cheng et al. 2008; Alumai et al. 2009). Lush, green monoculture lawns result from frequent irrigation, herbicide application, and the use of time-release fertilizers (Nielson and Smith 2005).
More than half of US households apply synthetic fertilizers to their yards, but the variability in application rate, timing and amount is high (Robbins et al. 2001; Law et al.
2004; Osmond and Hardy 2004). Do-it-yourself approaches result in a significant range of fertilizer application rates (e.g., 10-370 kg N ha-1 yr-1; Robbins et al. 2001; Law et al.
2004; Osmond and Hardy 2004). In contrast, professional management services have a smaller range of application rates (e.g. 100-161 kg N ha-1 yr-1), but apply with greater frequency compared to residents (Osmond and Platt 2000; Law et al. 2004). Overall, fertilizer use can be predicted by yard greenness (Zhou et al. 2008). When asked, residents report adjusting fertilizer application rates based on soil tests, land-cover type, season, and product instructions (Osmond and Hardy 2004). However, interviews from
Oregon reveal that only 20% read packaging labels, and the remainder “just know” the appropriate application rates or learn from their family and friends (Nielson and Smith
30 2005). In national and local surveys of residents, nearly half returned grass clippings to the yard, which reduces the need for fertilizers (NGA 2004; Osmond and Hardy 2004).
Nearly 75% of US households use chemical yard pesticides, including herbicides, insecticides and fungicides (Kiely et al. 2004). Similarly, across four North Carolina cities, an average of 60% of households (range 35–91%) used pesticides (Osmond and
Hardy 2004). Both residents and professional landscapers apply pesticides to control insects (i.e. insecticides) and weeds (i.e. herbicides) more than fungicides (Braman et al.
1997; Kiely et al. 2004; Osmond and Hardy 2004). Like fertilizer, pesticide use is related to groundcover, as a Phoenix study found that insecticides and herbicides were used less on grass yards than rock yards (Larson et al. 2010). Thus, xeric yards present potential tradeoffs between water use and environmental toxicity, which could negatively affect human health (e.g., Karr et al. 2007), beneficial urban organisms, or downstream water quality (e.g., Blanchoud et al. 2004; Struger and Fletcher 2007). Integrated pest management offers opportunities to reduce these tradeoffs and the environmental and economic consequences (Braman et al. 2000; Klingeman et al. 2009; Alumai et al. 2010).
Watering practices depend on yard cover, irrigation technology and decisions about how much and when to water. Physical yard features such as turfgrass, lot area, and house area are positive predictors of water use for pools and irrigation, which together constitute the majority of municipal and residential water consumption (Table 4; Mayer et al. 1999; Troy and Holloway 2004; Sovocool et al. 2006; Balling and Gober 2007;
Guhathakurta and Gober 2007; Wentz and Gober 2007; Harlan et al. 2009). Highly manicured lawns use significantly more water than the average yard (Askew and
McGuirk 2004), while xeric yards require less water, maintenance costs, and time
31 (Sovocool et al. 2006). However, over-watering is endemic in many yards (Nielson and
Smith 2005; Salvador et al. 2011). In somewhat counterintuitive findings, households with automated irrigation systems, including relatively efficient drip-irrigation, tend to use more water than households with hand-held or movable sprinklers (Mayer et al. 1999;
Martin 2001; Syme et al. 2004; Endter-Wada et al. 2008). Automated systems are often programmed at high rates, regardless of season and the needs of plants (Martin 2001). A recent survey found that regional water shortages are more likely to drive homeowners to conserve outdoor water than price increases or environmental concerns (St Hilaire et al.
2010). Yard water conservation efforts should target changing groundcover, species composition, irrigation technology, and homeowner education (Kjelgren et al. 2000).
Multi-scalar human drivers of residential landscape attributes and management
Numerous social theories and studies inform our understanding of the driving forces affecting landscape management and, in turn, the ecological outcomes of human behavior. Within the residential landscape literature, drivers of human behavior have been studied primarily at the household-scale (74% of 84 studies on human drivers), despite myriad factors across multiple scales that influence management practices and preferences of homeowners. These factors include personal attributes, neighborhood institutions, government policies, and broad-scale political-economic forces (Fig. 2;
Table 4). Across the studies reviewed here, attitudinal factors (e.g. environmental values) and household characteristics (e.g. demographics) have been most closely examined as the drivers of individual land management decisions. Overall, residents’ values and attitudes have limited influence on landscaping practices, in part because institutional and
32 structural forces facilitate and constrain people’s choices at household, neighborhoods and larger scales.
Household-scale human drivers: Residents are the fundamental local actors making landscaping decisions in front and backyards, where residents’ individual attitudes and social characteristics influence yard preferences and management practices. Following
Stern (2000), we distinguish between the cognitive realm of residents’ values and attitudinal judgments versus other factors (i.e., context, habits and personal attributes) that facilitate or constrain decisions (Fig. 2e, Household Scale Human Drivers). In our framework, cognitive factors encompass attitudes and related judgments, such as values, beliefs, and norms, while household and urban structure involves personal and property attributes such as wealth and housing age. Overall, attitudinal factors influence landscaping preferences and behaviors in complex but limited ways, with specific beliefs and attitudes (e.g., about one’s own conservation practices relative to others) affecting landscaping practices more so than general judgments (e.g., broad-based environmental worldviews; Larson et al. 2010). As further detailed below, we also find from our review that household and property attributes, such as income or property size, appear to impose stronger constraints on landscaping decisions and ecological characteristics than attitudinal or cognitive factors at the household-scale (Table 4).
Residents’ landscape preferences, or intentional desires, are not always in agreement with realized yard choices, particularly in public front yards (Hurd 2006; Larsen and
Harlan 2006). This disconnect, attributed to broader institutional and structural constraints that inhibit preferences from being realized, is less significant in private backyards where preferences are more closely aligned with actual landscapes (Larsen and
33 Harlan 2006). Front and backyard preferences differed in a Phoenix survey, along with the rationales for residents’ choices across these public and private spheres (Larson et al.
2009a; see also Goffman 1959 and Stern 2000 for a discussion of public versus private- shere actions). Thus, front yards may reflect a display of social status or adherence to neighborhood norms or rules, whereas backyards reflect residents’ ideals or
“dreamscapes” based on personal values and lifestyles (Larsen and Harlan 2006).
In qualitative studies, residents’ stated reasons for their landscaping preferences reflect value-based priorities. For example, aesthetic preferences are often a top priority in explaining groundcover and management choices (Martin et al. 2003; Spinti et al.
2004; Nielson and Smith 2005; Hirsch and Baxter 2009), in addition to the familiarity of landscapes, microclimate effects, and health factors (Larson et al. 2009a). Resident preferences for well-manicured versus natural or “messy” looking alternative landscapes vary based on aesthetic, safety, and environmental concerns (Jorgensen et al. 2007;
Mustafa et al. 2010; Zheng et al. 2011). Residents invoke environmental concerns as a landscaping priority, yet their perceptions of the ecological outcomes of various landscapes are mixed. Many people choose xeric landscaping for water conservation, while environmental benefits such as air quality are linked to grass and other yard types
(Larson et al. 2009a). Additionally, grassy yards are chosen as safe, comfortable, and
“homey” places for leisure activities, especially in backyards. Landscaping choices often reflect their utilitarian value for leisure, especially to provide recreational opportunities, minimize maintenance requirements, and address safety concerns (Harlan et al. 2006;
Clayton 2007; Endter-Wada et al. 2008; Yabiku et al. 2008; Hirsch and Baxter 2009;
Larson et al. 2009a). Yard choices also reflect non-utilitarian values concerning family
34 and social priorities such as feeling proud about one’s yard (Feagan and Ripmeester
1999; Endter-Wada et al. 2008; Hirsch and Baxter 2009). As such, landscapes represent symbolic expressions of residents’ identity (Larsen and Harlan 2006; Mustafa et al.
2010), while reflecting personal and social ideals based on what people think is important and how they subjectively view the world around them (Larson et al. 2009a). Thus, existing landscaping may influence people’s decisions to purchase a particular home or manage their yard in a certain way.
As broad-based notions of what is important in life (Schwartz 1994), values influence specific beliefs and attitudes about how the world works and what goals and actions are most desirable (Whittaker et al. 2006). Attitudinal constructs, which encompass a range of concerns, worldviews, and other types of evaluative (positive or negative) judgments, have been conceptualized and measured in myriad and often ambiguous ways, complicating comparisons across studies (Kaiser et al. 1999; Dunlap and Jones 2002;
Dietz et al. 2005). However, past research that quantitatively examines attitudinal factors and landscaping choices indicates complex relationships between environmental values and ecologically friendly landscaping practices (Larson et al. 2010). For example, pro- environmental values—measured by Dunlap’s New Ecological Paradigm scale and
“concern” about water scarcity—do not always translate into preferences for water- conserving xeric landscapes (Yabiku et al. 2008). Similarly, environmentally oriented people and those with heightened environmental “concern” tend to manage their yards more intensively than others, specifically with conventional chemical inputs (Templeton et al. 1999; Robbins et al. 2001; Robbins and Birkenholtz 2003). These counterintuitive
35 findings are consistent with the social construction of nature, wherein residents view green lawns as “nature,” even in a desert environment (Larson et al. 2009a).
Other landscape management practices, such as water use, are also influenced by personal beliefs, attitudes, and overriding social pressures in complex ways. Outdoor water use is often attributed to the belief that yards contribute to a property’s economic value (Nielson and Smith 2005; Endter-Wada et al. 2008). However, the link between water use and other attitudinal judgments is mixed. For example, preferences for a lush, green landscape explained outdoor water use in Australia (Askew and McGuirk 2004), but not in arid regions of the US like Utah and Arizona (Endter-Wada et al. 2008; Harlan et al. 2009). Positive attitudes toward water conservation practices explained reduced water use in Australia (Syme et al. 2004), while in a Mexican study, the belief that neighbors use water wastefully diminished conservation in outdoor irrigation practices
(Corral-Verdugo et al. 2002; Corral-Verdugo et al. 2003). Differences across studies may be attributed to the geographic context of the research or the particular types of value-based judgments examined, with specific beliefs and attitudes about behaviors being more influential than broad-based values or concerns.
As a specific type of belief, normative views about how people should maintain their yard to meet others’ expectations also influence environmental behaviors (Ajzen 1985;
Stern 2000). Few studies of residential landscaping practices empirically examine norms.
In the few studies available, residents often explain their management practices by citing neighborly expectations and pressures (Nielson and Smith 2005) or referencing their neighbors’ preferences more so than broad social norms (e.g., maintaining a monocultural lawn; Nassauer et al. 2009). Similarly, the desire for social acceptance was positively
36 related to the desire to conserve water, but did not translate to reduced water use (Syme et al. 2004). This finding highlights the tenuous relationship between what people say and what they do.
In addition to attitudinal drivers, personal attributes, interests and abilities of human actors drive landscape characteristics and management practices at the household-scale.
Income, for instance, is positively associated with vegetation cover (e.g., Mennis 2006;
Boone et al. 2010). This positive relationship is described by ecologists as a “luxury effect” resulting from the financial ability to create ecologically-rich landscapes, and by social scientists as a “prestige effect” involving symbolic displays of identity and social status beyond economic wherewithal (Martin et al. 2004; Kinzig et al. 2005; Grove et al.
2006b; Hope et al. 2006; Troy et al. 2007). Yet, financial resources, as well as time constraints, influence management choices by restricting the capacity to modify landscapes (Templeton et al. 1999; Hurd et al. 2006; Boone et al. 2010). Specifically, income predicts landscape preferences (Larsen and Harlan 2006), irrigation time, and outdoor water consumption (Osmond and Hardy 2004; Sovocool et al. 2006; Harlan et al.
2009; Polebitski and Palmer 2010). Middle-income residents, however, tend to prefer grass (Larsen and Harlan 2006) and apply the most fertilizers (Osmond and Platt 2000;
Law et al. 2004). On the other hand, wealthy residents and homeowners more often use pesticides than others (Steer et al. 2006; Templeton et al. 2008) and are willing to pay more for locally grown native plants (Curtis and Cowee 2010).
Finally, other personal and property attributes further shape management decisions.
In Baltimore, lawn care expenditures are positively related to income, but also predicted by education, median house value, and home ownership (Zhou et al. 2009). In the
37 Southwestern US, long-term residents, those with young children, and women prefer lawns more than xeric landscapes (Martin et al. 2003; Spinti et al. 2004; Larsen and
Harlan 2006; Yabiku et al. 2008; Larson et al. 2009a). Gender differences are attributed to the socialized duties of women as primary housekeepers and caregivers of children, while men commonly manage the yard and thus prefer more maintenance-free landscape types (Yabiku et al. 2008; Larson et al. 2009a). Property values, another indicator of affluence and social status, are also a good predictor of water consumption (Sovocool et al. 2006) and chemical use (Templeton et al. 1999; Robbins et al. 2001; Robbins and
Sharp 2003a). Finally, housing age is commonly related to particular front yard landscape types (Larsen and Harlan 2006) and is positively associated to vegetation cover
(Grove et al. 2006a; Grove et al. 2006b).
Neighborhood scale human drivers: Neighborhoods are fundamental units that shape landscape patterns within cities, due in part to developments with similar characteristics and in part to social institutions that foster or inhibit landscaping decisions both formally or informally (Feagan and Ripmeester 1999; Fig. 2f, Neighborhood Scale Human
Drivers). As organizing mechanisms for human actions, institutions are the rules, norms, and shared strategies that endure through social organization and interaction (Crawford and Ostrom 1995). Formal institutions include rules and restrictions codified in legal policies, while informal institutions encompass shared norms and non-codified codes of conduct. These factors partially explain the divergence between front and backyard choices, with social institutions operating to maintain the collective interests of residents in visible outdoor areas while constraining individuals’ choices (Table 4).
38 Formal institutions at the neighborhood scale drive resident landscaping practices through direct governance mechanisms. For example, homeowner association covenants, codes and restrictions (HOA CCRs), which are more prevalent in the US than other parts of the world, stipulate what residents can and cannot do to their homes and yards
(McKenzie 1994). As private governing bodies, HOAs and similar organizations enforce pre-existing land-use regulations, approve (or deny) landscape changes, collect dues, maintain common areas, and aim to uphold property values and “community standards”
(McKenzie 1994; Robbins 2007; Cheshire et al. 2009). While HOAs have the potential to regulate ecosystem services and water conservation, CCRs rarely govern outdoor water use explicitly but do regulate irrigation technology and restrict plant composition and height (Dyckman 2008). Yet the particular stipulations vary across neighborhoods; for example, some CCRs prohibit grass lawns while others require them in addition regulating pest and weed management. In Phoenix, HOA neighborhoods had fewer trees, more shrubs, and less turfgrass than non-HOA neighborhoods (Martin et al. 2003). These patterns are also likely linked to housing age, as developments with HOAs tend to be younger subdivisions (Larsen and Harlan 2006).
Although informal institutions, such as norms and customs, are not legally enforceable, the threat of social exclusion (or other informal penalties) shapes individual decisions to behave within widely accepted or perceived expectations. For example, the traditional lawn is ubiquitous in urban yards, partly as a result of neighborhood peer pressure and social conformity (Jenkins 1994; Askew and McGuirk 2004; Steinberg
2006; Robbins 2007). Well-kept lawns are commonly thought to reflect positively on the character, standards and value of the neighborhood (Feagan and Ripmeester 1999;
39 Robbins 2007). Most residential landscape studies offer only theoretical explanations for these social pressures, with anecdotal or minimal evidence. Few empirical studies have focused on understanding landscaping norms and how they operate across individuals, neighborhoods, or distinct regions. Together, formal and informal institutions can inhibit individuals from realizing their personal preferences and ideals, but they also explain why actual landscaping choices or yard structure deviate from residents’ values or attitudes.
Municipal and broader-scale drivers: At broader-scales, institutional and political- economic factors regulate landscaping decisions, enabling or constraining individuals’ choices (Fig. 2g, Municipal-Regional Scale Human Drivers). Despite their importance, limited research has focused on the broad-scale drivers of yard management, such as formal government ordinances and key political-economic players like the yard-care industry or large-scale housing developers (43% of 84 human driver studies).
Additionally, less is known about the relationship between residents’ decisions and broad-scale informal institutions, such as shared customs and notions about a region’s identity (Table 4).
Similar to HOAs, local government ordinances and restrictions formally control lawn maintenance or front yard structure. For example, some planning and development codes restrict the hard surfacing (paving) of yards or new driveways to reduce local flooding
(Stone 2004; Perry and Nawaz 2008; Shaffer et al. 2009). Municipal regulations often also dictate landscape structure or appearance through legal limits on grass height to avoid “nuisance” yards (Feagan and Ripmeester 1999; Robbins et al. 2001; Robbins
2007). Recent regulations in Canada banned cosmetic use of pesticides (Sandberg and
40 Foster 2005) in an attempt to address environmental concerns related to lawn care. For similar reasons, water-use restrictions in the Boston area now limit irrigation of lawns and gardens (Hill and Polsky 2007). Curiously, while relatively humid regions such as
Boston, Massachusetts and Miami, Florida have implemented outdoor irrigation restrictions, arid cities such as Phoenix lack regulations on residential water use even in a ten-year drought (Larson et al. 2009b). As Hill and Polsky (2007) explain, municipal decisions to regulate water are a function of political will and tolerance for water supply versus demand, rather than simple responses to hydroclimatic conditions.
Marketing and other political-economic factors reinforce the notion of a “perfect” lawn to promote happy homes and idyllic communities (Robbins et al. 2001). The commercial production of yard chemicals during post-war mid-1900s codified the ability to maintain the ideal American landscape and, thus, promoted the commodification of yards (Jackson 1985; Jenkins 1994; Bormann et al. 2001; Robbins and Sharp 2003a).
Adhering to industry standards, residents are entrenched in the marketing of chemical and yard-care corporations that perpetuate the consumption of lawns (Jenkins 1994; Bormann et al. 2001; Robbins and Sharp 2003a; Robbins 2007) and alternative landscapes, despite their perceived lower inputs (Mustafa et al. 2010). To alter this cycle, neighborhood institutions and municipal controls may be necessary to break residents from social pressures and economic forces (Robbins 2007; Nassauer et al. 2009).
Developers’ decisions also structure landscaping patterns and regulations that maintain neighborhood and regional characteristics (Coiacetto 2007; Cheshire et al.
2009). The development industry has powerful influence over broad-scale social- ecological outcomes, with similar building and greening strategies for master planned
41 communities and the ability to enact CCRs in new subdivisions (McGuirk and Dowling
2007; Dyckman 2008; Song et al. 2009). Many residents in fact prefer the master planned communities for their aesthetic uniformity and well-manicured landscapes, as well as a sense of social distinction (Dowling et al. 2010). Although driven mostly by profits, developers are influenced and constrained by consumer preferences, as well as municipal ordinances, past land uses and infrastructure, water and material costs, and the potential ecosystem services provided by yards (Coiacetto 2007; He and Jia 2007;
Dyckman 2008; Bowman and Thompson 2009; Mohamed 2009).
Finally, shared regional identities influence residents’ landscaping ideals and preferences. Place-based identity and the “sense of place” may vary over time, leading to
“cultural landscapes” reflecting shared customs, developers’ original decisions, and residents’ ideals (Gobster et al. 2007; Romig 2010). For example, preferences for lush yards by long-time Phoenix residents are influenced by the promotion of Phoenix as an
“oasis”—where “the desert is a myth” (Larson et al. 2009b). In contrast, residents of
Tucson, Arizona embrace tourism campaigns that promote the desert setting (Prytherch
2002), where xeric yards are more prevalent and rates of water use are lower than
Phoenix. Thus, the sense of place promoted in particular areas may have lasting effects by socializing residents to cultural landscaping practices and ideals (Larson et al. 2009a).
Legacy effects
Landscape legacies manifest at multiple scales through historic land-use decisions, preexisting land-cover, and urban development patterns. Legacies result from numerous social and ecological forces, ultimately affecting ecological structure, function and
42 services for centuries to millennia (Redman 1999; Foster et al. 2003; Lewis et al. 2006;
Fig. 2h, Legacy Effects). Few studies, however, highlight the strong influence of historic trajectories on residential land management (4% of all 256 studies). The landscaping choices and consumption patterns of previous residents, which may have been driven by local climate, culture or available supplies, often create long-lasting legacies (Dow 2000;
Alp et al. 2010; Boone et al. 2010). Recent studies suggest neighborhood land cover is better predicted by previous, rather than current, socio-economic and lifestyle characteristics in neighborhoods (Luck et al. 2009; Boone et al. 2010). Further, previous land-cover decisions continue to shape cultural expectations, for example, to maintain traditional lawn yards in historic neighborhoods of Phoenix (Larson et al. 2009a).
At larger scales, ecological legacies are a feature of the prevailing decisions of developers and previously established infrastructure or land uses. Developers’ original landscaping and infrastructure choices (e.g., irrigation technology) and the use of conventional versus alternative design principles may have long-lasting effects on regional characteristics that go unchanged due to cost constraints or CCR regulations
(Larsen and Harlan 2006; Dyckman 2008; Conway 2009). Historic and present-day access to infrastructure also impacts landscape structure and management practices. As is the case of many Phoenix neighborhoods, water rights and canals used for early crop irrigation were easily transitioned into flood irrigation systems that now maintain lush lawns in the desert (Gober and Trapido-Lurie 2006). Similarly, yards built on previously farmed land in Phoenix have fewer woody species (Hope et al. 2003) and greater soil nutrients than those built on non-agrarian, desert lands (Lewis et al. 2006).
43 DISCUSSION AND FUTURE RESEARCH DIRECTIONS
Residential landscapes are an ideal setting to examine human (resident)-environment
(yard) interactions. Residential landscapes research can integrate disciplinary perspectives across a common system, while serving as a model for studying social- ecological complexity in other systems (Baker et al. 2007). From over 250 papers published on residential landscapes (Fig. 1), most early studies focused on identifying the components of the residential landscape system (Fig. 2) from theory-driven strengths within social and natural science disciplines, often at single scales of analysis. As the field has matured, interdisciplinary studies have become more common, addressing the interactions between people and their surrounding home environment using mixed methodologies (Fig. 1). Still, further research is needed to understand interactions between the drivers of land management decisions and social or ecological outcomes across multiple scales. Interdisciplinary research can especially elucidate the tradeoffs and multi-scalar dynamics associated with human-environment systems.
Conceptual frameworks are useful tools for illustrating the links between disparate disciplinary perspectives or research in a common, interconnected system. Our approach builds upon existing urban ecosystem frameworks (e.g., Grimm et al. 2000; Pickett et al.
2001; Redman et al. 2004) by applying the causes, consequences, and feedbacks specific to residential land management to the broad components of previous social-ecological frameworks. While recent interdisciplinary frameworks have explored specific components of urban systems, such as public green spaces (James et al. 2009) and residential yard management at the parcel scale (Byrne and Grewal 2009), our framework
44 distinctively adds an essential cross-scalar perspective that considers the parcel scale as well as neighborhood and broader, regional dynamics of landscaping decisions. In particular, the framework uniquely highlights the multiple scales of social drivers that influence household decisions regarding their outdoor space. The framework herein is useful in pinpointing areas for future disciplinary and integrated research for social- ecological systems broadly and residential landscapes specifically, as further described below and in Table 5. Although specific to residents’ land management decisions, our framework could be adapted and refined for other social-ecological systems or contexts, such as public outdoor spaces managed by government officials.
Future research on residential landscapes will benefit from forging interdisciplinary theories with a common vocabulary and integrative approaches (Table 5). Disciplinary framing of research questions with varying theories and epistemologies is a commonly encountered barrier to effective collaboration on complex systems such as residential landscapes (Eigenbrode et al. 2007; Miller et al. 2008). For example, while social scientists recognize that residents create their own version of “nature”—both mentally and materially—in their well-manicured yards (i.e., following theories about the social construction of nature), ecologists traditionally consider a “natural landscape” to more closely represent the properties and functioning of the undisturbed, surrounding native ecosystem (Feagan and Ripmeester 1999). Nevertheless, this barrier can be overcome if scientists work collaboratively outside of their disciplinary training. Integrative approaches with strong conceptual foundations, including concepts like ecosystem services that bridge ecological functions with the values people ascribe to them and conceptual frameworks such as the one presented here, will facilitate this work.
45 In addition to developing integrated theoretical foundations, research is also needed on understudied system components (boxes Fig. 2) and interactions (arrows Fig. 2) of the residential landscape system. Ecologically, most studies are limited by their focus on ecosystem properties in front yards, but would benefit from examining how ecological properties and processes vary across public (front) and private (backyard) spaces within residential parcels (Table 5). For example, few ecological studies have examined backyards, or distinguish between front and backyards. This is in part due to available methods that rely on from-the-street observations or remote sensing, which coarsely classifies land use and land cover. Social science research, however, has identified variation in the drivers of landscaping practices of the public and private spheres of typical single-family residences. Thus, front and backyards are likely to have distinct landscapes and management practices, and in turn, ecological patterns and processes.
Beyond parcels, the neighborhood to regional ecology of residential landscapes largely remains unknown, particularly in terms of broad-scale ecosystem functioning and services (Table 5). The heterogeneous urban matrix and diversity of management choices at the household-scale make regional generalizations difficult. Further exploration is needed to identify the potential tradeoffs between the ecological outcomes of management practices at all scales, particularly those that lead to ecosystem services and disservices (Table 5). Future research should also link multi-scalar remote sensing technology with ground-level investigations of ecological functioning and social drivers to broaden understanding of cross-scalar feedbacks between the drivers of management practices and their ecological outcomes.
46 Many household-scale ecological studies note the importance of landscape management practices, yet few quantitatively account for variability in water and chemical inputs. The logistical constraints associated with replication and managing individual resident’s behaviors across multiple study plots make quantifying specific practices difficult. However, empirical research (e.g., Law et al. 2004) suggests chemical inputs vary not only among neighbors, but also between do-it-yourself and professional yard care managers. By generalizing yard care based on professional standards and not accounting for residents’ actual management practices, ecologists cannot accurately calculate variation in ecological properties and functions at the parcel-scale. Inaccurate generalizations about yard management also hinder realistic predictions of ecological processes at broader-scales. Controlled experiments can test the relationship between management practices and plot-scale variables or variables that are easily assessed at larger scales like lawn greenness or microclimate. Integrating social science methods, such as resident surveys about management practices, with ecological field studies will be essential to quantitatively capture the impacts of disparate landscaping practices and understand the heterogeneity of ecological characteristics among parcels (Table 5).
Similarly, modeling tools such as the Household Flux Calculator (Baker et al. 2007) will help to identify the distribution and outcomes of various management strategies across households and cities.
In the social sciences, the human drivers of management decisions are relatively well studied, particularly at the household-scale. However, the variety of social constructs, research methods, and cross-scalar feedbacks result in mixed findings and are difficult to generalize (Table 4). For example, due to the disparate ways in which cognitive
47 judgments are conceived, the magnitude of and interaction between attitudinal factors and landscape management remains unclear. Future research can begin to clarify conflicting findings through consistent conceptualizations of human drivers, particularly cognitive factors, and the use of common methods and definitions (Table 5). Additionally, the relationship between individual agent-based drivers and broad-scale structural drivers of landscaping decisions is unclear. While the field of ecology routinely examines biophysical processes at multiple scales (Wu et al. 2006), many social scientists tend to favor a particular scale or unit of analysis (e.g., households in Larsen and Harlan 2005 versus the broader political economy in Robbins and Sharp 2003b). Social theories about the multi-scalar drivers of human decision-making are therefore underdeveloped, especially in relation to land management and resulting ecological patterns and processes.
This is mainly due to the lack of studies examining drivers across scales and different socio-political contexts or locations. Multi-scalar and cross-site research that takes advantage of mixed methodologies, both qualitative and quantitative, will assist in clarifying the multiple influences on human decisions and in explaining disconnects between attitudes, preferences, and behaviors across varying contexts.
Legacy effects from previous land-use/land-cover and management decisions have not been studied extensively in residential landscapes, but they influence social- ecological outcomes and current or future landscaping decisions. A historical perspective provides important context for feedbacks that operate not only across temporal and spatial scales, but also between humans and the environment. Future research will benefit from documenting former land cover, land use, and social context as well as understanding the temporal or spatial patterns that emerge from these (Table 5; Dow
48 2000). For example, if individuals’ preferences for water-conserving landscapes grow, developers and large-scale market economies may begin to accommodate these new preferences. While shifts in broader institutions are often slow, changes in these institutions may ultimately feedback to affect resident’s management decisions and choice of home based on the existing landscape. In the future, these choices will be a legacy of past decisions, playing a role in how landscapes are managed over time.
Finally, systematic comparisons of social and ecological patterns and interactions are needed across diverse cities and scales using comparable methods (Table 5). Most previous studies have taken a case-specific approach focused on parcel- or household- scale research questions in a single region. Narrow focus on a single geographic context may conceal patterns that diverge in different biophysical (e.g., climatic or physiographic) and social (e.g., political or demographic) contexts. Cross-site research will reveal how and what human-environment patterns and processes emerge in distinct geographies or across diverse settings. Future cross-site research will thereby contribute to generalizable knowledge as well as nuanced, contextual understanding of complex social-ecological interactions that are place-specific.
CONCLUSIONS
Our synthesis of the growing body of literature on residential landscape social- ecology reveals a number of interesting trends and future research opportunities. While a great deal of research has focused on the parcel-household-scale in a single city, the social drivers and ecological characteristics of residential landscapes in fact vary across
49 scales, from distinct landscape patches within parcels (e.g. front versus backyards), to households, neighborhoods and regions. Additionally, research efforts to date have focused primarily on the individual components of the system, such as the ecological structure of yards or social factors that affect human choices (i.e. the boxes in Fig. 2), and far less is known about the interactions and tradeoffs that occur among these components
(i.e. the arrows in Fig. 2). The social and ecological components of the residential landscape system intersect most clearly within two key but understudied areas, the link between drivers and outcomes of management decisions made by people, and the ecosystem services provided by these landscapes to residents (Table 5). In addition, patterns and feedbacks that emerge across scales, as well as across multiple sites and regions, have not been extensively studied and are not well understood.
The synthesis and framework we present draws on insights that emerged from the cumulative social, ecological and interdisciplinary literature on residential landscapes.
The suggested integrated conceptual approach brings together disparate, disciplinary research to clarify patterns and dynamics between system components. In addition, this framework advances integrative urban social-ecological theory and research. Residential landscapes are a dominant land use within urban systems, and their management at parcel and broader-scales will have important implications for human well-being, urban ecological functioning, and the continued provision of ecosystem services. Thus, understanding the feedbacks and tradeoffs between human drivers, ecological outcomes and ecosystem services must also coincide with planning and management strategies at multiple scales that maximize benefits to both people and the environment.
50 ACKNOWLEDGEMENTS
This material is based upon work supported by the National Science Foundation (NSF) under Grant No. DEB-0423704, Central Arizona–Phoenix Long-Term Ecological
Research (CAP LTER) and Grant No. 0504248, Integrative Graduate Education Research and Training (IGERT) in Urban Ecology. We thank B. Funke, C.A Strawhacker, and
V.K. Turner for insightful discussions on the conceptual diagram presented in this paper, which was conceived during the 2008 IGERT / ASU School of Sustainability workshop.
We also thank two anonymous reviewers for their helpful comments that improved this manuscript.
51 REFERENCES
Abraham A, Sommerhalder K, Abel T (2010) Landscape and well-being: A scoping study on the health-promoting impact of outdoor environments. Int J Public Health 55 (1):59-69
Acar C, Acar H, Eroglu E (2007) Evaluation of ornamental plant resources to urban biodiversity and cultural changing: A case study of residential landscapes in Trabzon City (Turkey). Build Environ 42 (1):218-229
Ajzen I (1985) From intentions to actions: A theory of planned behavior. In: Kuhl J, Beckmann J (eds) Action control, from cognition to behavior. Springer series in social psychology. Springer-Verlag, New York, pp 11-39
Akinnifesi F, Sileshi G, Ajayi O, Akinnifesi A, de Moura E, Linhares J, Rodrigues I (2010) Biodiversity of the urban homegardens of São Luís City, northeastern Brazil. Urban Ecosyst 13 (1):129-146
Albuquerque UP, Andrade LHC, Caballero J (2005) Structure and floristics of homegardens in northeastern Brazil. J Arid Environ 62 (3):491-506
Alp S, Ozturk S, Turkoglu N, Koyuncu M (2010) Basic elements of the traditional garden identity in the city of Van. Afr J Agric Res 5 (11):1277-1283
Alumai A, Grunkemeyer M, Kovach J, Shetlar D, Cardina J, Rimelspach J, Clayton S, Grewal P (2010) Implementing integrated pest management in professional lawn care: A case study. Urban Ecosyst 13 (1):37-49
Alumai A, Salminen S, Richmond D, Cardina J, Grewal P (2009) Comparative evaluation of aesthetic, biological, and economic effectiveness of different lawn management programs. Urban Ecosyst 12 (2):127-144
Andersson E, Barthel S, Ahrne K (2007) Measuring social-ecological dynamics behind the generation of ecosystem services. Ecol Appl 17 (5):1267-1278
Askew LE, McGuirk PM (2004) Watering the suburbs: Distinction, conformity and the suburban garden. Aust Geographer 35 (1):17-37
Aurora A, Simpson T, Small M, Bender K (2009) Toward increasing avian diversity: Urban wildscapes programs. Urban Ecosyst 12 (3):347-358
52 Baker LA, Hartzheim PM, Hobbie SE, King JY, Nelson KC (2007) Effect of consumption choices on fluxes of carbon, nitrogen and phosphorus through households. Urban Ecosyst 10 (2):97-117
Baker PJ, Harris S (2007) Urban mammals:What does the future hold? An analysis of the factors affecting patterns of use of residential gardens in Great Britain. Mamm Rev 37 (4):297-315
Balling RC, Gober P (2007) Climate variability and residential water use in the city of Phoenix, Arizona. J Appl Meteorol Climatol 46 (7):1130-1137
Band LE, Cadenasso ML, Grimmond CS, Grove JM, Pickett STA (2005) Heterogeneity in urban ecosystems: Patterns and process. In: Lovett GM, Turner MG, Jones CG, Weathers KC (eds) Ecosystem function in heterogeneous landscapes. Springer, New York, pp 257-278
Baris ME, Sahin S, Yazgan ME (2009) The contribution of trees and green spaces to the urban climate: The case of Ankara. Afr J of Agric Res 4 (9):791-800
Beard JB, Green RL (1994) The role of turfgrass in environemntal protection and their benefits to humans. J Environ Qual 23 (3)
Bennett EM, Carpenter SR, Clayton MK (2005) Soil phosphorus variability: Scale- dependence in an urbanizing agricultural landscape. Landsc Ecol 20 (4):389-400
Bhatti M, Church A (2001) Cultivating natures: Homes and gardens in late modernity. Sociol 35 (02):365-383
Bijoor NS, Czimczik CI, Pataki DE, Billings SA (2008) Effects of temperature and fertilization on nitrogen cycling and community composition of an urban lawn. Glob Change Biol 14 (9):2119-2131
Blanchoud H, Farrugia F, Mouchel JM (2004) Pesticide uses and transfers in urbanised catchments. Chemosph 55 (6):905-913
Bonan GB (2000) The microclimates of a suburban Colorado (USA) landscape and implications for planning and design. Landsc Urban Plan 49 (3-4):97-114
Boone CG, Cadenasso ML, Grove JM, Schwarz K, Buckley GL (2010) Landscape, vegetation characteristics, and group identity in an urban and suburban watershed: Why the 60s matter. Urban Ecosyst 13:255-271
Bormann FH, Balmori D, Geballe GT (2001) Redesigning the American lawn: A search for environmental harmony. 2nd edition. Yale University Press, New Haven, CT
53 Bowman T, Thompson J (2009) Barriers to implementation of low-impact and conservation subdivision design: Developer perceptions and resident demand. Landsc Urban Plan 92 (2):96-105
Braman SK, Latimer JG, Oetting RD, McQueen RD, Eckberg TB, Prinster M (2000) Management strategy, shade, and landscape composition effects on urban landscape plant quality and arthropod abundance. J Econ Entomol 93 (5):1464- 1472
Braman SK, Oetting RD, Florkowski W (1997) Assessment of pesticide use by commercial landscape maintenance and lawn care firms in Georgia. J Entomol Sci 32 (4):403-411
Burghardt KT, Tallamy DW, Shriver WG (2009) Impact of native plants on bird and butterfly biodiversity in suburban landscapes. Conserv Biol 23 (1):219-224
Buyantuyev A, Wu JG (2010) Urban heat islands and landscape heterogeneity: Linking spatiotemporal variations in surface temperatures to land-cover and socioeconomic patterns. Landsc Ecol 25:17-33
Byrne LB (2006) Effects of of urban habitat types and landscape patterns on ecological variables at the aboveground–belowground interface. PhD Dissertation, Pennsylvania State University
Byrne LB (2007) Habitat structure: A fundamental concept and framework for urban soil ecology. Urban Ecosyst 10 (3):255-274
Byrne LB, Bruns MA (2004) The effects of lawn management on soil microarthropods. J Agr Urban Entomol 21 (3):150-156
Byrne LB, Bruns MA, Kim KC (2008) Ecosystem properties of urban land covers at the aboveground-belowground interface. Ecosyst 11 (7):1065-1077
Byrne LB, Grewal P (2008) Introduction to ecological landscaping: A holistic description and framework to guide the study and management of urban landscape parcels. CATE 1 (2)
Catterall CP, Cousin JA, Piper S, Johnson G (2010) Long-term dynamics of bird diversity in forest and suburb: Decay, turnover or homogenization? Divers Distrib 16 (4):559-570
Chace JF, Walsh JJ (2006) Urban effects on native avifauna: A review. Landsc Urban Plan 74 (1):46-69
54 Chamberlain DE, Cannon AR, Toms MP (2004) Associations of garden birds with gradients in garden habitat and local habitat. Ecography 27 (5):589-600
Cheng Z, Richmond D, Salminen S, Grewal P (2008) Ecology of urban lawns under three common management programs. Urban Ecosyst 11 (2):177-195
Cheshire LA, Rosenblatt T, Lawrence G, Walters P (2009) The governmentality of master planning: Housing consumption, aesthetics and community on a new estate. Hous Stud 24 (5):653-667
Clayton S (2007) Domesticated nature: Motivations for gardening and perceptions of environmental impact. J Environ Psychol 27 (3):215-224
Coiacetto E (2007) The role of the development industry in shaping urban social space: A conceptual model. Geog Res 45 (4):340-347
Collins JP, Kinzig A, Grimm NB, Fagan WF, Hope D, Wu JG, Borer ET (2000) A new urban ecology. Am Sci 88 (5):416-425
Conway T (2009) Local environmental impacts of alternative forms of residential development. Environ Plan B-Plan Des 36 (5):927-943
Cook WM, Faeth SH (2006) Irrigation and land use drive ground arthropod community patterns in an urban desert. Environ Entomol 35 (6):1532-1540
Corral-Verdugo V, Frias-Amenta M, Gonzalez-Lomeli D (2003) On the relationship between anti-social and anti-environmental behaviors: An empirical study. Popul Environ 24 (3):273-286
Corral-Verdugo V, Frias-Armenta M, Perez-Urias F, Orduna-Cabrera V, Espinoza- Gallego N (2002) Residential water consumption, motivation for conserving water and the continuing tragedy of the commons. Environ Manag 30 (4):527-535
Coutts AM, Beringer J, Tapper NJ (2007) Impact of increasing urban density on local climate: Spatial and temporal variations in the surface energy balance in Melbourne, Australia. J of Appl Meteorol Climatol 46 (4):477-493
Crawford SES, Ostrom E (1995) A grammar of institutions. Am Political Sci Rev 89 (3):582-600
Crow T, Brown T, De Young R (2006) The Riverside and Berwyn experience: Contrasts in landscape structure, perceptions of the urban landscape, and their effects on people. Landsc Urban Plan 75 (3-4):282-299
55 Curtis KR, Cowee MW (2010) Are homeowners willing to pay for “origin-certified" plants in water-conserving residential landscaping? J Agric Resour Econ 35 (1):118-132
Daniels GD, Kirkpatrick JB (2006a) Comparing the characteristics of front and back domestic gardens in Hobart, Tasmania, Australia. Landsc Urban Plan 78 (4):344- 352
Daniels GD, Kirkpatrick JB (2006b) Does variation in garden characteristics influence the conservation of birds in suburbia? Biol Conserv 133 (3):326-335
Davies R, Hall SJ (2010) Direct and indirect effects of urbanization on soil and plant nutrients in desert ecosystems of the Phoenix metropolitan area, Arizona (USA). Urban Ecosyst 13 (3):295-317
Davies ZG, Fuller RA, Loram A, Irvine KN, Sims V, Gaston KJ (2009) A national scale inventory of resource provision for biodiversity within domestic gardens. Biol Conserv 142 (4):761-771
Dietz T, Fitzgerald A, Shwom R (2005) Environmental values. Annu Rev Environ Resour 30:335-372
Doody BJ, Sullivan JJ, Meurk CD, Stewart GH, Perkins HC (2010) Urban realities: The contribution of residential gardens to the conservation of urban forest remnants. Biodivers Conserv 19 (5):1385-1400
Dorney JR, Guntenspergen GR, Keough JR, Stearns F (1984) Composition and structure of an urban woody plant community. Urban Ecol 8 (1-2):69-90
Dow K (2000) Social dimensions of gradients in urban ecosystems. Urban Ecosyst 4 (4):255-275
Dowling R, Atkinson R, McGuirk P (2010) Privatism, privatisation and social distinction in master-planned residential estates. Urban Policy Res 28 (4):391-410
Dunlap RE, Jones RE (2002) Environmental conern: Conceptual and measurement issues. In: Dunlap RE, Michelson W (eds) Handbook of Environmental Sociology. Greenwood Press, Westport, CT, pp 482-542
Dyckman CS (2008) The covenant conundrum in urban water conservation. Urban Lawyer 40 (1):17-50
Edgar DR, Kershaw GP (1994) The density and diversity of the bird populations in three residential communities in Edmonton, Alberta. Can Field-Nat 108 (2):156-161
56 Effland WR, Pouyat RV (1997) The genesis, classification, and mapping of soils in urban areas. Urban Ecosyst 1 (4):217-228
Eichemberg MT, Amorozo MCD, de Moura LC (2009) Species composition and plant use in old urban homegardens in Rio Claro, southeast of Brazil. Acta Bot Bras 23 (4):1057-1075
Eigenbrode SD, O'Rourke M, Wulfhorst JD, Althoff DM, Goldberg CS, Merrill K, Morse W, Nielsen-Pincus M, Stephens J, Winowiecki L, Bosque-Perez NA (2007) Employing philosophical dialogue in collaborative science. Biosci 57 (1):55-64
Endter-Wada J, Kurtzman J, Keenan SP, Kjelgren RK, Neale CMU (2008) Situational waste in landscape watering: Residential and business water use in an urban Utah community. J Am Water Resour Assoc 44 (4):902-920
Erickson JE, Cisar JL, Volin JC, Snyder GH (2001) Comparing nitrogen runoff and leaching between newly established St. Augustine grass turf and an alternative residential landscape. Crop Sci 41 (6):1889-1895
Evans KL, Newson SE, Gaston KJ (2009) Habitat influences on urban avian assemblages. Ibis 151 (1):19-39
Faeth SH, Warren PS, Shochat E, Marussich WA (2005) Trophic dynamics in urban communities. Biosci 55 (5):399-407
Feagan RB, Ripmeester M (1999) Contesting natural(ized) lawns: A geography of private green space in the Niagara region. Urban Geogr 20 (7):617-634
Fetridge ED, Ascher JS, Langellotto GA (2008) The bee fauna of residential gardens in a suburb of New York City (Hymenoptera: Apoidea). Ann Entomol Soc Am 101 (6):1067-1077
Foster D, Swanson F, Aber J, Burke I, Brokaw N, Tilman D, Knapp A (2003) The importance of land-use legacies to ecology and conservation. Biosci 53 (1):77-88
French K, Major R, Hely K (2005) Use of native and exotic garden plants by suburban nectarivorous birds. Biol Conserv 121 (4):545-559
Fuller RA, Irvine KN, Devine-Wright P, Warren PH, Gaston KJ (2007) Psychological benefits of greenspace increase with biodiversity. Biol Lett 3 (4):390-394
Fuller RA, Warren PH, Armsworth PR, Barbosa O, Gaston KJ (2008) Garden bird feeding predicts the structure of urban avian assemblages. Divers Distrib 14 (1):131-137
57 Gaston KJ, Warren PH, Thompson K, Smith RM (2005) Urban domestic gardens (IV): The extent of the resource and its associated features. Biodivers Conserv 14 (14):3327-3349
Gavier-Pizarro GI, Radeloff VC, Stewart SI, Huebner CD, Keuler NS (2010) Housing is positively associated with invasive exotic plant species richness in new england, USA. Ecol Appl 20 (7):1913-1925
Georgi JN, Dimitriou D (2010) The contribution of urban green spaces to the improvement of environment in cities: Case study of Chania, Greece. Build Environ 45 (6):1401-1414
Germaine SS, Rosenstock SS, Schweinsburg RE, Richardson WS (1998) Relationships among breeding birds, habitat, and residential development in greater Tucson, Arizona. Ecol Appl 8 (3):680-691
Germaine SS, Schweinsburg RE, Germaine HL (2001) Effects of residential density on Sonoran desert nocturnal rodents. Urban Ecosyst 5 (3):179-185
Gober P, Trapido-Lurie B (2006) Metropolitan Phoenix: Place making and community building in the desert. Metropolitan Portraits. University of Pennsylvania Press, Philadelphia, PA
Gobster PH, Nassauer JI, Daniel TC, Fry G (2007) The shared landscape: What does aesthetics have to do with ecology? Landsc Ecol 22 (7):959-972
Goddard MA, Dougill AJ, Benton TG (2010) Scaling up from gardens: Biodiversity conservation in urban environments. Trends Ecol Evol 25 (2):90-98
Goffman E (1959) The presentation of self in everyday life. Doubleday Anchor Books. Doubleday, Garden City, NY
Golubiewski NE (2006) Urbanization increases grassland carbon pools: Effects of landscaping in Colorado's front range. Ecol Appl 16 (2):555-571
Green DM, Oleksyszyn M (2002) Enzyme activities and carbon dioxide flux in a Sonoran desert urban ecosystem. Soil Sci Soc of Am J 66:2002-2008
Grimm NB, Faeth SH, Golubiewski NE, Redman CL, Wu JG, Bai XM, Briggs JM (2008) Global change and the ecology of cities. Sci 319 (5864):756-760
Grimm NB, Grove JM, Pickett STA, Redman CL (2000) Integrated approaches to long- term studies of urban ecological systems. Biosci 50 (7):571-584
58 Groffman PM, Law NL, Belt KT, Band LE, Fisher GT (2004) Nitrogen fluxes and retention in urban watershed ecosystems. Ecosyst 7 (4):393-403
Groffman PM, Pouyat RV (2009) Methane uptake in urban forests and lawns. Environ Sci Technol 43 (14):5229-5235
Groffman PM, Williams CO, Pouyat RV, Band LE, Yesilonis ID (2009) Nitrate leaching and nitrous oxide flux in urban forests and grasslands. J Environ Qual 38 (5):1848-1860
Grove JM, Cadenasso ML, Burch WR, Pickett STA, Schwarz K, O'Neil-Dunne J, Wilson M, Troy A, Boone C (2006a) Data and methods comparing social structure and vegetation structure of urban neighborhoods in Baltimore, Maryland. Soc Nat Resour 19 (2):117-136
Grove JM, Troy AR, O'Neil-Dunne JPM, Burch WR, Cadenasso ML, Pickett STA (2006b) Characterization of households and its implications for the vegetation of urban ecosystems. Ecosyst 9 (4):578-597
Guhathakurta S, Gober P (2007) The impact of the Phoenix urban heat island on residential water use. J Am Plan Assoc 73 (3):317-329
Hall SJ, Ahmed B, Ortiz P, Davies R, Sponseller RA, Grimm NB (2009) Urbanization alters soil microbial functioning in the Sonoran desert. Ecosyst 12 (4):654-671
Hall SJ, Huber D, Grimm NB (2008) Soil N2O and NO emissions from an arid, urban ecosystem. J Geophys Res-Biogeosci 113 (G1)
Hall T (2010) Goodbye to the backyard?: The minimisation of private open space in the Australian outer-suburban estate. Urban Policy Res 28 (4):411-433
Harlan SL, Brazel AJ, Prashad L, Stefanov WL, Larsen L (2006) Neighborhood microclimates and vulnerability to heat stress. Soc Sci Med 63 (11):2847-2863
Harlan SL, Yabiku ST, Larsen L, Brazel AJ (2009) Household water consumption in an arid city: Affluence, affordance, and attitudes. Soc Nat Resour 22 (8):691-709
He JJ, Jia B (2007) Sustainable residential landscapes: A case study in Guangzhou, China. Lands Res 32 (2):241-254
Henderson SPB, Perkins NH, Nelischer M (1998) Residential lawn alternatives: A study of their distribution, form and structure. Landsc Urban Plan 42 (2-4):135-145
Hill TD, Polsky C (2007) Suburbanization and drought: A mixed methods vulnerability assessment in rainy Massachusetts. Environ Hazards 7 (4):291-301
59
Hirsch R, Baxter J (2009) The look of the lawn: Pesticide policy preference and health- risk perception in context. Environ Plan C-Gov Policy 27 (3):468-490
Holland JH (1995) Hidden order: How adaptation builds complexity. Helix Books. Addison-Wesley, Reading, MA
Hope D, Gries C, Casagrande D, Redman CL, Grimm NB, Martin C (2006) Drivers of spatial variation in plant diversity across the Central Arizona-Phoenix ecosystem. Soc Nat Resour 19 (2):101-116
Hope D, Gries C, Zhu WX, L. FW, Redman CL, Grimm NB, Nelson AL, Martin C, Kinzig AP (2003) Socioeconomics drive urban plant diversity. PNAS 100 (15):8788-8792
Hope D, Zhu W, Gries C, Oleson J, Kaye J, Grimm NB, Baker L (2005) Spatial variation in soil inorganic nitrogen across an arid urban ecosystem. Urban Ecosyst 8 (3):251-273
Hourigan CL, Catterall CP, Jones D, Rhodes M (2010) The diversity of insectivorous bat assemblages among habitats within a subtropical urban landscape. Austral Ecol 35 (8):849-857
Huai H, Hamilton A (2009) Characteristics and functions of traditional homegardens: A review. Front Biol Chin 4 (2):151-157
Huang YJ, Akbari H, Taha H, Rosenfeld AH (1987) The potential of vegetation in reducing summer cooling loads in residential buildings. J Clim Appl Meteorol 26 (9):1103-1116
Hurd BH (2006) Water conservation and residential landscapes: Household preferences, household choices. J Agric Resour Econ 31 (2):173-192
Hurd BH, St Hilaire R, White JM (2006) Residential landscapes, homeowner attitudes, and water-wise choices in New Mexico. HortTech 16 (2):241-246
Iverson LR, Cook EA (2000) Urban forest cover of the Chicago region and its relation to household density and income. Urban Ecosyst 4:104-124
Jackson KT (1985) Crabgrass frontier: The suburbanization of the United States. Oxford University Press, NY
James P, Tzoulas K, Adams MD, et al. (2009) Towards an integrated understanding of green space in the European built environment. Urban For Urban Green 8 (2):65- 75
60
Jenerette GD, Harlan SL, Brazel A, Jones N, Larsen L, Stefanov WL (2007) Regional relationships between surface temperature, vegetation, and human settlement in a rapidly urbanizing ecosystem. Landsc Ecol 22 (3):353-365
Jenerette GD, Wu JG, Grimm NB, Hope D (2006) Points, patches, and regions: Scaling soil biogeochemical patterns in an urbanized arid ecosystem. Glob Change Biol 12 (8):1532-1544
Jenkins VS (1994) The lawn: A history of an American obsession. Smithsonian Institution Press, Washington, DC
Jenny H (1941) Factors of soil formation: A system of quantitative pedology. McGraw- Hill, New York, London
Jorgensen A, Hitchmough J, Dunnett N (2007) Woodland as a setting for housing appreciation and fear and the contribution to residential satisfaction and place identity in Warrington New Town, UK. Landsc Urban Plan 79:273–28
Kabir ME, Webb EL (2009) Household and homegarden characteristics in southwestern Bangladesh. Agrofor Syst 75 (2):129-145
Kaiser FG, Wolfing S, Fuhrer U (1999) Environmental attitude and ecological behaviour. J Environ Psychol 19 (1):1-19
Karr CJ, Solomon GM, Brock-Utne AC (2007) Health effects of common home, lawn, and garden pesticides. Pediatr Clin N Am 54 (1):63-80
Kaye JP, Burke IC, Mosier AR, Guerschman JP (2004) Methane and nitrous oxide fluxes from urban soils to the atmosphere. Ecol Appl 14 (4):975-981
Kaye JP, Majumdar A, Gries C, Buyantuyev A, Grimm NB, Hope D, Jenerette GD, Zhu WX, Baker L (2008) Hierarchical bayesian scaling of soil properties across urban, agricultural, and desert ecosystems. Ecol Appl 18 (1):132-145
Kaye JP, McCulley RL, Burke IC (2005) Carbon fluxes, nitrogen cycling, and soil microbial communities in adjacent urban, native and agricultural ecosystems. Glob Change Biol 11 (4):575-587
Kiely T, Donaldson D, Grube A (2004) Pesticide industry sales and usage: 2000 and 2001 market estimates. Environmental Protection Agency; Office of Pesticide Programs, Washington DC
61 Kinzig AP, Warren P, Martin C, Hope D, Katti M (2005) The effects of human socioeconomic status and cultural characteristics on urban patterns of biodiversity. Ecol Soc 10 (1):23
Kirkpatrick J, Daniels G, Davison A (2009) An antipodean test of spatial contagion in front garden character. Landsc Urban Plan 93 (2):103-110
Kirkpatrick J, Daniels G, Zagorski T (2007) Explaining variation in front gardens between suburbs of Hobart, Tasmania, Australia. Landsc Urban Plan 79 (3- 4):314-322
Kjelgren R, Rupp L, Kilgren D (2000) Water conservation in urban landscapes. Hortsci 35 (6):1037-1040
Klingeman WE, Pettis GV, Braman SK (2009) Lawn care and landscape maintenance: Professional acceptance of insect- and disease-resistant ornamental plants. Hortsci 44 (6):1608-1615
Koerner B, Klopatek J (2002) Anthropogenic and natural CO2 emission sources in an arid urban environment. Environ Pollut 116:S45-S51
Kweon BS, Ellis CD, Leiva PI, Rogers GO (2010) Landscape components, land use, and neighborhood satisfaction. Environ Plan B-Plan Des 37 (3):500-517
Larsen L, Harlan SL (2006) Desert dreamscapes: Residential landscape preference and behavior. Landsc Urban Plan 78 (1-2):85-100
Larson KL, Casagrande D, Harlan SL, Yabiku ST (2009a) Residents' yard choices and rationales in a desert city: Social priorities, ecological impacts, and decision tradeoffs. Environ Manag 44 (5):921-937
Larson KL, Cook EM, Strawhacker CS, Hall SJ (2010) The influence of diverse values, ecological structure and geographic context on residents’ multifaceted landscaping decisions. Hum Ecol 38 (6):747-761
Larson KL, Gustafson A, Hirt P (2009b) Insatiable thirst and a finite supply: An assessment of municipal water-conservation policy in greater Phoenix, Arizona, 1980-2007. J Policy Hist 21 (02):107-137
Law NL, Band LE, Grove JM (2004) Nitrogen input from residential lawn care practices in suburban watersheds in Baltimore County, MD. J Environ Plan Manag 47 (5):737-755
Lehmann A, Stahr K (2007) Nature and significance of anthropogenic urban soils. J Soils Sediments 7 (4):247-260
62
Lewis DB, Kaye JP, Gries C, Kinzig AP, Redman CL (2006) Agrarian legacy in soil nutrient pools of urbanizing arid lands. Glob Change Biol 12:703-709
Liu HB, Hull RJ, Duff DT (1997) Comparing cultivars of three cool-season turfgrasses - for soil water NO3 concentration and leaching potential. Crop Sci 37 (2):526-534
Livesley SJ, Dougherty BJ, Smith AJ, Navaud D, Wylie LJ, Arndt SK (2010) Soil- atmosphere exchange of carbon dioxide, methane and nitrous oxide in urban garden systems: Impact of irrigation, fertiliser and mulch. Urban Ecosyst 13 (3):273-293
Livingston M, Shaw WW, Harris LK (2003) A model for assessing wildlife habitats in urban landscapes of eastern Pima County, Arizona (USA). Landsc Urban Plan 64:131-144
Loram A, Thompson K, Warren PH, Gaston KJ (2008a) Urban doemestic gardens (XII): The richness and composition of the flora in five UK cities. J Veg Sci 19 (3):321- 330
Loram A, Tratalos J, Warren PH, Gaston KJ (2007) Urban domestic gardens (X): The extent & structure of the resource in five major cities. Landsc Ecol 22 (4):601-615
Loram A, Warren PH, Gaston KJ (2008b) Urban domestic gardens (XIV): The characteristics of gardens in five cities. Environ Manag 42 (3):361-376
Loss SR, Ruiz MO, Brawn JD (2009) Relationships between avian diversity, neighborhood age, income, and environmental characteristics of an urban landscape. Biol Conserv 142 (11):2578-2585
Luck GW, Smallbone LT, O'Brien R (2009) Socio-economics and vegetation change in urban ecosystems: Patterns in space and time. Ecosyst 12 (4):604-620
MacArthur RH, Wilson EO (1967) The theory of island biogeography. Princeton University Press, Princeton, NJ
Marco A, Dutoit T, Deschamps-Cottin M, Mauffrey JF, Vennetier M, Bertaudiere- Montes V (2008) Gardens in urbanizing rural areas reveal an unexpected floral diversity related to housing density. Comptes Rendus Biologies 331 (6):452-465
Martin CA (2001) Landscape water use in Phoenix, Arizona. Desert Plants 17:26-31
Martin CA (2008) Landscape sustainability in a Sonoran desert city. CATE 1 (2)
63 Martin CA, Busse K, Yabiku S (2007) North Desert Village: The effect of landscape manipulation on microclimate and its relation to human landscape preference. Hortsci 42 (4):853-854
Martin CA, Peterson KA, Stabler LB (2003) Residential landscaping in Phoenix, Arizona, USA: Practices and preferences relative to covenants, codes and restrictions. J Arboric 29 (1):9-17
Martin CA, Warren PS, Kinzig AP (2004) Neighborhood socioeconomic status is a useful predictor of perennial landscape vegetation in residential neighborhoods and embedded small parks of Phoenix, AZ. Landsc Urban Plan 69 (4):355-368
Marzluff JM, Ewing K (2001) Restoration of fragmented landscapes for the conservation of birds: A general framework and specific recommendations for urbanizing landscapes. Restor Ecol 9 (3):280-292
Mathieu R, Freeman C, Aryal J (2007) Mapping private gardens in urban areas using object-oriented techniques and very high-resolution satellite imagery. Landsc Urban Plan 81 (3):179-192
Matson PA, Naylor R, Ortiz-Monasterio I (1998) Integration of environmental, agronomic, and economic aspects of fertilizer management. Sci 280 (5360):112- 115
Matsuoka RH, Kaplan R (2008) People needs in the urban landscape: Analysis of landscape and urban planning contributions. Landsc Urban Plan 84 (1):7-19
Mayer PW, DeOreo WB, et al. (1999) Residential end uses of water. American Water Works Association Research Foundation, Denver, CO
McGuirk PM, Dowling R (2007) Understanding master-planned estates in Australian cities: A framework for research. Urban Policy Res 25 (1):21-38
McIntyre NE, Hostetler ME (2001) Effects of urban land use on pollinator (Hymenoptera: Apoidea) communities in a desert metropolis. Basic Appl Ecol 2 (3):209-218
McKenzie E (1994) Privatopia: Homeowner associations and the rise of residential private government. Yale University Press, New Haven, CT
McKinney ML (2006) Urbanization as a major cause of biotic homogenization. Biol Conserv 127 (3):247-260
MEA (2005) Ecosystems and human well-being: Synthesis. Millenium Ecosystem Assessment. Island Press, Washington, DC
64
Mennis J (2006) Socioeconomic-vegetation relationships in urban, residential land: The case of Denver, Colorado. Photogramm Eng Remote Sens 72 (8):911-921
Milesi C, Running SW, Elvidge CD, Dietz JB, Tuttle BT, Nemani RR (2005) Mapping and modeling the biogeochemical cycling of turf grasses in the United States. Environ Manag 36 (3):426-438
Mittelbach GG, Steiner CF, Scheiner SM, Gross KL, Reynolds HL, Waide RB, Willig MR, Dodson SI, Gough L (2001) What is the observed relationship between species richness and productivity? Ecol 82 (9):2381-2396
Mohamed R (2009) Why do residential developers prefer large exurban lots? Infrastructure costs and exurban development. Environ Plan B-Plan Des 36 (1):12-29
Mustafa D, Smucker TA, Ginn F, Johns R, Connely S (2010) Xeriscape people and the cultural politics of turfgrass transformation. Environ Plan D-Soc Space 28 (4):600-617
Nassauer JI, Wang ZF, Dayrell E (2009) What will the neighbors think? Cultural norms and ecological design. Landsc Urban Plan 92 (3-4):282-292
NGA (2004) Environmental lawn and garden survey. National Gardening Association, South Burlington, VT
NGA (2007) Residential lawn and landscape services and the value of landscaping. National Gardening Association, South Burlington, Vermont
Nielson L, Smith CL (2005) Influences on residential yard care and water quality: Tualatin watershed, Oregon. J Am Water Resour Assoc 41 (1):93-106
Nowak DJ, Rowntree RA, McPherson EG, Sisinni SM, Kerkmann ER, Stevens JC (1996) Measuring and analyzing urban tree cover. Landsc Urban Plan 36 (1):49-57
Oki L, Greco S, Gan J, Yates M, Haver D (2007) Characterizing the runoff from single- family residential landscapes. Hortsci 42 (4):911-911
Osmond DL, Hardy DH (2004) Characterization of turf practices in five North Carolina communities. J Environ Qual 33 (2):565-575
Osmond DL, Platt J (2000) Characterization of suburban nitrogen fertilizer and water use on residential turf in Cary, North Carolina. HortTechnol 10 (2):320-325
65 Perry T, Nawaz R (2008) An investigation into the extent and impacts of hard surfacing of domestic gardens in an area of Leeds, United Kingdom. Landsc Urban Plan 86 (1):1-13
Peters EB, McFadden JP (2010) Influence of seasonality and vegetation type on suburban microclimates. Urban Ecosyst 13:443-460.
Petrovic AM (1990) The fate of nitrogenous fertilizers applied to turfgrass. J Environ Qual 19 (1):1-14
Pickett STA, Burch WR, Dalton SE, Foresman TW, Grove JM, Rowntree R (1997) A conceptual framework for the study of human ecosystems in urban areas. Urban Ecosyst 1 (4):185-199
Pickett STA, Cadenasso ML (2009) Altered resources, disturbance, and heterogeneity: A framework for comparing urban and non-urban soils. Urban Ecosyst 12 (1):23-44
Pickett STA, Cadenasso ML, Grove JM, Groffman PM, Band LE, Boone CG, Burch WR, Grimmond CSB, Hom J, Jenkins JC, Law NL, Nilon CH, Pouyat RV, Szlavecz K, Warren PS, Wilson MA (2008) Beyond urban legends: An emerging framework of urban ecology, as illustrated by the Baltimore Ecosystem Study. Biosci 58 (2):139-150
Polebitski AS, Palmer RN (2010) Seasonal residential water demand forecasting for census tracts. J Water Resour Plan Manag-Asce 136 (1):27-36
Pouyat RV, Groffman PM, Yesilonis ID, Hernandez L (2002) Soil carbon pools and fluxes in urban ecosystems. Environ Pollut S107-S118
Pouyat RV, Yesilonis ID, Golubiewski NE (2009) A comparison of soil organic carbon stocks between residential turfgrass and native soil. Urban Ecosyst 12 (1):45-62
Pouyat RV, Yesilonis ID, Russell-Anelli J, Neerchal NK (2007) Soil chemical and physical properties that differentiate urban land-use and cover types. Soil Sci Soc Am J 71 (3):1010-1019
Prytherch DL (2002) Selling the eco-entrepreneurial city: Natural wonders and urban stratagems in Tucson, Arizona. Urban Geogr 23 (8):771-793
Qian YL, Bandaranayake W, Parton WJ, Mecham B, Harivandi MA, Mosier AR (2003) Long-term effects of clipping and nitrogen management in turfgrass on soil organic carbon and nitrogen dynamics: The CENTURY model simulation. J Environ Qual 32:1694-1700
66 Raciti SM, Groffman PM, Fahey TJ (2008) Nitrogen retention in urban lawns and forests. Ecol Appl 18 (7):1615-1626
Raupp MJ, Shrewsbury PM, Herms DA (2010) Ecology of herbivorous arthropods in urban landscapes. Ann Rev Entomol 55:19-38
Redman CL (1999) Human impact on ancient environments. University of Arizona Press, Tucson, AZ
Redman CL, Grove JM, Kuby LH (2004) Integrating social science into the Long-Term Ecological Research (LTER) network: Social dimensions of ecological change and ecological dimensions of social change. Ecosyst 7:161-171
Richards NA, Mallette JR, Simpson RJ, Macie EA (1984) Residential greenspace and vegetation in a mature city - Syracuse, New York. Urban Ecol 8 (1-2):99-125
Robbins P (2007) Lawn people: How grasses, weeds, and chemicals make us who we are. Temple University Press, Philadelphia, PA
Robbins P, Birkenholtz T (2003) Turfgrass revolution: Measuring the expansion of the American lawn. Land Use Policy 20 (2):181-194
Robbins P, Polderman A, Birkenholtz T (2001) Lawns and toxins: An ecology of the city. Cities 18 (6):369-380
Robbins P, Sharp JT (2003a) Producing and consuming chemicals: The moral economy of the American lawn. Econ Geogr 79 (4):425-451
Robbins P, Sharp JT (2003b) The lawn-chemical economy and its discontents. Antipode 35 (5):955-979
Romig K (2010) Community and social capital in upper-income neighborhoods: An investigation in metropolitan Phoenix. Urban Geogr 31 (8):1065-1079
Ryznar RM, Wagner TW (2001) Using remotely sensed imagery to detect urban change – viewing Detroit from space. J Am Plan Assoc 67 (3):327-336
Salvador R, Bautista-Capetillo C, Playan E (2011) Irrigiation performance in private urban landscapes: A study case in Zaragoza (Spain). Landsc Urban Plan 100:302- 311
Sandberg LA, Foster J (2005) Challenging lawn and order: Environmental discourse and lawn care reform in Canada. Environ Politics 14 (4):478-494
67 Scharenbroch BC, Lloyd JE, Johnson-Maynard JL (2005) Distinguishing urban soils with physical, chemical, and biological properties. Pedobiologia 49:283-296
Schwartz S (1994) Are there universal aspects in the structure and contents of human values? J Soc Issues 50 (4):19-45
Shaffer P, Wilson S, Brindle F, Baffoe-Bonnie, B (2009) Understanding permeable and impermeable surfaces: Technical report on surfacing options and cost benefit analysis. Department of Communities and Local Government, UK. www.communities.gov.uk/ publications/planningandbuilding/permeablesufacesreport. Accessed 30 Jan 2011
Shashua-Bar L, Pearlmutter D, Erell E (2009) The cooling efficiency of urban landscape strategies in a hot dry climate. Landsc Urban Plan 92 (3-4):179-186
Shochat E, Lerman SB, Anderies JM, Warren PS, Faeth SH, Nilon CH (2010) Invasion, competition, and biodiversity loss in urban ecosystems. Biosci 60 (3):199-208
Shochat E, Stefanov WL, Whitehouse MEA, Faeth SH (2004) Urbanization and spider diversity: Influences of human modification of habitat structure and productivity. Ecol Appl 14 (1):268-280
Shochat E, Warren PS, Faeth SH, McIntyre NE, Hope D (2006) From patterns to emerging processes in mechanistic urban ecology. Trends Ecol Evol 21 (4):186- 191
Simpson JR, McPherson EG (1998) Simulation of tree shade impacts on residential energy use for space conditioning in Sacramento. Atmos Environ 32 (1):69-74
Smetak KM, Johnsn-Maynard JL, Lloyd JE (2007) Earthworm population density and diversity in different-aged urban systems. Appl Soil Ecol 37 (1-2):161-168
Smith J, Chapman A, Eggleton P (2006a) Baseline biodiversity surveys of the soil macrofauna of London’s green spaces. Urban Ecosyst 9 (4):337-349
Smith RM, Gaston KJ, Warren PH, Thompson K (2005) Urban domestic gardens (V): Relationships between landcover composition, housing and landscape. Landsc Ecol 20 (2):235-253
Smith RM, Gaston KJ, Warren PH, Thompson K (2006b) Urban domestic gardens (VIII): Environmental correlates of invertebrate abundance. Biodivers Conserv 15 (8):2515-2545
68 Smith RM, Thompson K, Hodgson JG, Warren PH, Gaston KJ (2006c) Urban domestic gardens (IX): Composition and richness of the vascular plant flora, and implications for native biodiversity. Biol Conserv 129 (3):312-322
Smith RM, Warren PH, Thompson K, Gaston KJ (2006d) Urban domestic gardens (VI): Environmental correlates of invertebrate species richness. Biodivers Conserv 15 (8):2415-2438
Song K, Qin J, Gao K, Hu YH (2009) Homogenization of plant diversity in Shanghai residential areas. Yingyong Shengtai Xuebao 20 (7):1603-1607
Sovocool KA, Morgan M, Bennett D (2006) An in-depth investigation of xeriscape as a water conservation measure. J Am Water Works Assoc 98 (2):82-93
Sperling C, Lortie C (2010) The importance of urban backgardens on plant and invertebrate recruitment: A field microcosm experiment. Urban Ecosyst 13 (2):223-235
Spinti JE, St Hilaire R, VanLeeuwen D (2004) Balancing landscape preferences and water conservation in a desert community. HortTechnol 14 (1):72-77
St Hilaire R, Arnold MA, Wilkerson DC, Devitt DA, Hurd BH, Lesikar BJ, Lohr VI, Martin CA, McDonald GV, Morris RL, Pittenger DR, Shaw DA, Zoldoske DF (2008) Efficient water use in residential urban landscapes. Hortsci 43 (7):2081- 2092
St Hilaire R, VanLeenwen DM, Torres P (2010) Landscape preferences and water conservation choices of residents in a high desert environment. HortTech 20 (2):308-314
Stabler LB (2008) Management regimes affect woody plant productivity and water use efficiency in an urban desert ecosystem. Urban Ecosyst 11:197-211
Stabler LB, Martin CA, Brazel AJ (2005) Microclimates in a desert city were related to land use and vegetation index. Urban For Urban Green 3 (3-4):137-147
Steer CD, Grey CNB, Team AS (2006) Socio-demographic characteristics of UK families using pesticides and weed-killers. J Expo Sci Environ Epidemiol 16 (3):251-263
Steinberg T (2006) American green: The obsessive quest for the perfect lawn. 1st edition. W.W. Norton, New York
Stern PC (2000) Toward a coherent theory of environmentally significant behavior. J Soc Issues 56 (3):407-424
69
Stewart GH, Meurk CD, Ignatieva ME, Buckley HL, Magueur A, Case BS, Hudson M, Parker M (2009) Urban biotopes of Aotearoa New Zealand (URBANZ) II: Floristics, biodiversity and conservation values of urban residential and public woodlands, Christchurch. Urban For Urban Green 8 (3):149-162
Stone JB (2004) Paving over paradise: How land use regulations promote residential imperviousness. Landsc Urban Plan 69 (1):101-113
Struger J, Fletcher T (2007) Occurrence of lawn care and agricultural pesticides in the Don River and Humber River watersheds (1998-2002). J Gt Lakes Res 33 (4):887-905
Su WZ, Gu CL, Yang GS (2010) Assessing the impact of land use/land cover on urban heat island pattern in Nanjing City, China. J Urban Plan Dev-ASCE 136 (4):365- 372
Syme GJ, Shao QX, Po M, Campbell E (2004) Predicting and understanding home garden water use. Landsc Urban Plan 68 (1):121-128
Templeton SR, Yoo SJ, Zilberman D (1999) An economic analysis of yard care and synthetic chemical use: The case of San Francisco. Environ Resour Econ 14 (3):385-397
Templeton SR, Zilberman D, Yoo SJ, Dabalen AL (2008) Household use of agricultural chemicals for soil-pest management and own labor for yard work. Environ Resour Econ 40 (1):91-108
Thompson K, Austin KC, Smith RM, Warren PH, Angold PG, Gaston KJ (2003) Urban domestic gardens (I): Putting small-scale plant diversity in context. J Veg Sci 14 (1):71-78
Thompson K, Hodgson JG, Smith RM, Warren PH, Gaston KJ (2004) Urban domestic gardens (III): Composition and diversity of lawn floras. J Veg Sci 15 (3):373-378
Tilman D, Cassman KG, Matson PA, Naylor R, Polasky S (2002) Agricultural sustainability and intensive production practices. Nature 418 (6898):671-677
Townsend-Small A, Czimczik CI (2010) Carbon sequestration and greenhouse gas emissions in urban turf. Geophys Res Lett 37: L02707
Tratalos J, Fuller RA, Evans KL, Davies RG, Newson SE, Greenwood JJD, Gaston KJ (2007a) Bird densities are associated with household densities. Glob Change Biol 13 (8):1685-1695
70 Tratalos J, Fuller RA, Warren PH, Davies RG, Gaston KJ (2007b) Urban form, biodiversity potential and ecosystem services. Landsc Urban Plan 83 (4):308-317
Troy AR, Grove JM, O'Neil-Dunne JPM, Pickett STA, Cadenasso ML (2007) Predicting opportunities for greening and patterns of vegetation on private urban lands. Environ Manag 40 (3):394-412
Troy P, Holloway D (2004) The use of residential water consumption as an urban planning tool: A pilot study in Adelaide. J Environ Plan Manag 47 (1):97-114
Trudgill S, Jeffery A, Parker J (2010) Climate change and the resilience of the domestic lawn. Appl Geogr 30 (1):177-190
Tzoulas K, Korpela K, Venn S, Yli-Pelkonen V, Kazmierczak A, Niemela J, James P (2007) Promoting ecosystem and human health in urban areas using green infrastructure: A literature review. Landsc Urban Plan 81 (3):167-178
UN (2010) World urbanization prospects: The 2009 revision. United Nations, NY
Uslu A, Gokce S (2010) Social interaction in urban transformation areas and the characteristics of urban outdoor spaces: A case study from Turkey. Afr J Agric Res 5 (20):2801-2810
Waide RB, Willig MR, Steiner CF, Mittelbach G, Gough L, Dodson SI, Juday GP, Parmenter R (1999) The relationship between productivity and species richness. Ann Rev Ecol Evol Syst 30:257-300
Wentz EA, Gober P (2007) Determinants of small-area water consumption for the city of Phoenix, Arizona. Water Resour Manage 21 (11):1849-1863
Whittaker D, Vaske JJ, Manfredo MJ (2006) Specificity and the cognitive hierarchy: Value orientations and the acceptability of urban wildlife management actions. Soc Nat Resour 19 (6):515-530
Wu J, Jones KB, Li H, Loucks OL (2006) Scaling and Uncertainty Analysis in Ecology: Methods and Applications. Springer, The Netherlands.
Yabiku ST, Casagrande DG, Farley-Metzger E (2008) Preferences for landscape choice in a Southwestern desert city. Environ Behav 40 (3):382-400
Yahner RH (2001) Butterfly communities in residential landscapes of central Pennsylvania. Northeast Nat 8 (1):113-118
Zheng B, Zhang YQ, Chen JQ (2011) Preference to home landscape: Wildness or neatness? Landsc Urban Plan 99 (1):1-8
71
Zhou WQ, Troy A, Grove JM, Jenkins JC (2009) Can money buy green? Demographic and socioeconomic predictors of lawn-care expenditures and lawn greenness in urban residential areas. Soc Nat Resour 22 (8):744-760
Zhou WQ, Troy A, Grove M (2008) Modeling residential lawn fertilization practices: Integrating high resolution remote sensing with socioeconomic data. Environ Manag 41 (5):742-752
Zmyslony J, Gagnon D (1998) Residential management of urban front yard landscape: A random process? Landsc Urban Plan 40 (4):295-307
Zmyslony J, Gagnon D (2000) Path analysis of spatial predictors of front yard landscape in an anthropogenic environment. Landsc Ecol 15 (4):357-371
72 been reported in the residential landscape l caratsfindings; indicate (^) the relationship has been reported at the neighborhood scale; and blank cells indicate no findi positive relationship; minus ( TABLE 1 :
Variables related to ecological properties residential of landscapes across the literature. (+) symbols indicate Plus a
Composition & Soil Properties Microclimate Composition trends may vary by species by mayregion. trends and species vary a Lands
Th Structure Faunal Cover Plant e relationshipsthe shown here dominant efromlandscapes however, literature; faunal trendsresidentialare composition cape cape Ecological property -
) symbols indicate
Relative Cooling Microclimate / phosphorus Bioavailable nitrogen Total & bioavailable Soil organic carbon Bulk density Bird species richness abundance Small mammal richness species Invertebrate abundance Invertebrate Cover Species richness features Area of biotic / abiotic
a
iterature a a
an inverse relationship; combined plus/minus ( + /
a
+/ + + + +
^ ^ ^ - Yard size
+ + ^ - - - - /
^ ^ ^ ^
-
Housing density
^
+ + + + ^ - + + -
/ Housing &
^ ^ ^ ^
-
^ neighborhood age factors explanatory Related
+ + + + ^ +
/ Socioeconomic &
^ ^ ^ -
^ lifestyle factors
+/
+ + + + + + + Management
-
practices +/
+ + + + + Plant composition &
-
structure
+ + + + + +
Former land use - +/
Habitat fragmentation - - ) symbols indicate mixed
^ ^ -
& distance to native ^ ecosystem
+ Regional climate &
^
urbanization patterns ngs havengs
73 TABLE 2: Variables most related to ecological functioning and ecosystem services of residential landscapes across the literature. Symbols as in Table 1
Related explanatory factors
Ecological Structure Management
e
matter
Ecological function Irrigation Ground & & Ground Time since since Time Fertilization disturbance Soil organic organic Soil composition temperatur Soil / surface surface / Soil Plant species species Plant and ecosystem services canopycover Water fluxes Evapotranspiration (Regulating service) + +/- + + Faunal interactions Trophic dynamics (Regulating service) + + + Methane (CH ) 4 consumption + - Carbon dioxide (CO2) Soil-plant-atmosphere emissions + + + + exchange and nutrient Carbon sequestration cycling (Regulating & + + + supporting services) Nitrous oxide (N O) 2 emissions +/- + + + - Nitrate (NO3 ) leaching +/- +/- - + + - Primary production + + Plant gas exchange + + Plant and faunal Pollination functions (Regulating, + supporting & Beneficial predation provisioning services) + Habitat provisions + + Subsistence provisions + Cooling + + + + Sense of place + + + Landscape functions Social cohesion (Cultural services) + + + Leisure and recreation + + + +
Property values + + + +
74 TABLE 3: Variables most related to residential landscape management decisions across the literature. Filled symbols () indicate variables are strongly related; open symbols () indicate mixed or an indirect relationship; plus (+) symbols indicate a positive relationship; blank cells indicate no findings have been reported
Related explanatory factors
ppropriate
monocultures Ground cover & & cover Ground Professional vs. vs. Professional yourself management yourself - vegetation structure vegetation it Irrigation technology
-
Landscape grass manicured, Highly Knowledge of a of Knowledge do Climate variability & season & variability Climate
management decisions etc. amount, rate, application Fertilization rate &
frequency ! " " +
Pesticide use (herbicides,
insecticides, etc.) " " +
Water use ! " + " "
75 literature. Symbols as in Table 3 4: TABLE
management Preferences
management practices practices management Landscape decisions appearances, which necessitate often these the management use of practices a Homeowner associations do not often directly regulate irrigation, pesticide, or fertilizer use; however, they do require cert and preferences
Variables most related to multi Landscape
expen Lawn care use Fertilizer use Pesticide Irrigation structure Yard structure Yard
diture
! ! ! !
Broad values
Attitudes & human !
Beliefs and attitudes cognition
! !
Preferences - scalar human residents’ drivers preferences of and management practices across the
" " " " "
Aesthetic tastes
! ! ! ! !
Environmental concern
! "
Education Social
! " " " " " Income
- Related Related structural characteristics " "
Gender
" "
Place of origin &
explanatory factors explanatory
length of residence " "
Home ownership
" " " "
Property value
" "
Property size & age
! ! ! ! i " nstitutions
Homeowners Formal
a a a a
association
" " " Municipal ordinances &
regulations institutions ! " " " Informal Informal Neighborhood norms
" " "
Social customs
ain yard " " " " economy Political Market dynamics
" " "
Development industry
76 TABLE 5: Future research opportunities within and across scales, geographic contexts, and disciplines to advance the study of human-environment interactions within residential landscape settings
Ecological properties & processes • Patterns across front/backyards, neighborhood, regional & larger scales • Comparable methods across studies • Ecological patterns across geographic contexts
Ecosystem services • Ecosystem services as an interdisciplinary concept & tool • Tradeoffs between ecosystem services & management practices • Dynamics of ecosystem services across scales
Management decisions • Explicit link between human drivers, management decisions, and ecological & social outcomes • Variability in landscaping decisions across contexts & scales
Multi-scalar human drivers • Clear, comparable attitudinal constructs & measurements • Social drivers linked to ecological outcomes across private & public spaces • Interactions between human drivers across scales • Role of institutions (formal & informal) & political-economic forces
Legacy effects • Link between social & ecological legacies & current or future decisions • Influence of historic land-use/land-cover & geographic context on present- day decisions
Integrated social-ecological research • Interdisciplinary social-ecological theory, concepts & methods applied across geographic contexts & scales
77 FIGURE 1: Number of publications (1980 – 2010, total N = 256) on social and ecological aspects of residential landscapes, categorized by natural science topics (n = 135), social science topics (n = 64), and those that are interdisciplinary (n = 57) in focus and methodology. Years 1980 - 1992 are grouped together because of the small number (8) of relevant literature published over the 12 year time period
78 FIGURE 2: Conceptual framework of multi-scalar social-ecological interactions of residential landscapes. Arrows represent interactions between framework components
79 CHAPTER 3
ATMOSPHERIC NITROGEN DEPOSITION IN A LARGE SEMI-ARID CITY IS
LOWER THAN EXPECTED: FINDINGS FROM A METHODS COMPARISON
ABSTRACT
Cities occupy a small land area globally, yet atmospheric compounds generated from human-dominated ecosystems have significant impacts on protected lands. Atmospheric nitrogen (N) deposition alters ecosystems, including biogeochemical cycling, primary production, and community composition. In arid ecosystems, considerable uncertainty surrounds estimates of atmospheric N inputs due to heterogeneous precipitation patterns and difficulties in quantifying dry deposition. I employed multiple approaches to quantify spatial and temporal patterns of N deposition at locations within Phoenix,
Arizona and the surrounding native desert. I compared N deposition measured by ion- exchange resin (IER) collectors (bulk and throughfall, 2006–2013), and N deposition estimated by the inferential method using passive samplers (atmospheric N concentrations x deposition velocity; 2010-2012). Over two summer and winter seasons,
I co-located samplers to directly compare methods for quantifying total and dry N deposition. Rates of total (wet and dry) N deposition estimated with throughfall collectors (3.3 kg N ha-1 yr-1) had minimal spatial variation and were significantly lower than expected based on model and other aridland estimates. Despite minimal spatial variation, N deposition varied temporally. For example, total N throughfall deposition was greatest in the summer monsoon season and best predicted by the frequency of
80 precipitation events. Contrary to throughfall estimates, inferential methods indicated elevated N deposition—in the form of ammonia, nitrogen oxides, and nitric acid deposition—was restricted to the urban core (6.4 kg N ha-1 yr-1). Overall, throughfall methods significantly underestimated total and dry deposition particularly in the urban region and in the winter season. By accounting for the potential underestimation by throughfall, I predicted average long-term N deposition in the region was 7.3 kg N ha-1 yr-1, which more closely approximated the modeled regional deposition but was still relatively low compared to other arid regions. Inconsistencies between approaches reveal how uncertainties related to quantifying dry deposition, as well as site characteristics and deposition velocities easily confound N deposition estimates. These findings highlight the need for and benefit of mixed methods to quantify wet and dry N deposition in arid systems. Overall, despite the size and population of arid Phoenix, N deposition was lower than expected compared to other cities, but was greatest in the urban core compared to the outlying desert.
81 INTRODUCTION
Increased human activities have led to elevated concentrations of atmospheric reactive nitrogen (N) worldwide (Vitousek et al. 1997; Galloway et al. 2004; Dentener et al. 2006). Emissions of oxidized and reduced N gases are transported downwind of urban and agricultural sources until they form secondary products or are removed from the atmosphere to surfaces through rainfall (wet deposition), cloud vapor, or adsorption of gases and particles (dry deposition). Rates of N deposition and the ecological effects have been well characterized in many ecosystems with high rainfall, including temperate and boreal forests, mesic grasslands, and heathlands (Lovett 1994; Aber et al. 1998;
Galloway et al. 2004; Holland et al. 2005; Weathers et al. 2006; Bobbink et al. 2010;
Pardo et al. 2011). In these regions, N deposition has increased during decades of urban and industrial growth and is projected to continue rising, with significant consequences for biogeochemical cycling and biotic composition (Phoenix et al. 2006; Dentener et al.
2006; Bobbink et al. 2010). Arid and semi-arid ecosystems cover over a third of the globe’s land area and future urban growth is expected to disproportionately occur in developing arid and semi-arid grassland and shrublands worldwide (UN 2009). Nitrogen deposition is thus expected to increase in dryland ecosystems, which likely respond differently than N-limited ecosystems where water does not constrain plant production.
Yet, the quantification of N deposition in dryland regions has received far less attention than in more temperate ecosystems. And, where aridland N deposition has been quantified, there are many uncertainties and challenges in quantifying dry deposition and inputs from spatially and temporally patchy precipitation events (Fenn et al. 2009).
82 Arid and semi-arid ecosystems are especially sensitive to small anthropogenic changes, such as elevated N (Fenn, Baron, et al. 2003; Pardo et al. 2011). Significant portions of aridland ecosystems are predicted to receive atmospheric N inputs at or above the desert critical load—the level of deposition between 3-8 kg N ha-1 yr-1 at which ecological changes occur (Fenn et al. 2010; Pardo et al. 2011). For example, in the semi- arid southwestern and western United States (US), dry deposition alone is estimated within the range of the critical load and as high as 14 - 35 kg N ha-1 yr-1 (Alonso,
Bytnerowicz, and Boarman 2005; L. E. Rao et al. 2009; Cisneros et al. 2010). Nitrogen inputs above the critical load in aridlands result in increased annual herbaceous plant growth, a loss of native desert vegetation, increased fire frequency, and lake eutrophication (Brooks 2003; Fenn, Baron, et al. 2003; Baron 2006; Báez et al. 2007; L.
E. Rao, Allen, and Meixner 2010; Hall et al. 2011). The low N critical load in drylands highlights the importance of accurately assessing total atmospheric inputs to arid systems and capturing the pulsed wet deposition and dry deposition over long time periods. In order to better predict when and where deposition will occur for more effective management and protection of native ecosystems, a better understanding of the patterns and drivers of N deposition is needed.
Atmospheric N inputs are expected to be particularly high in arid regions influenced by urbanization compared to more remote arid lands. For example, empirical estimates of N deposition range between 30 - 60 kg N ha-1 yr-1 in the metropolitan regions of Los
Angeles, California and the large Chinese cities of Urumqi and Beijing (Bytnerowitz and
Fenn 1996; Fenn et al. 2003; Pan et al. 2012; Li et al. 2013). Yet, in the sixth largest city in the US, estimates of N deposition in Phoenix, Arizona and the surrounding Sonoran
83 Desert are an order of magnitude lower (3-6 kg N ha-1 yr-1) than other arid urban regions and than regional models predict (Baker et al. 2001; Fenn, Haeuber, et al. 2003; Lohse et al. 2008). It is uncertain if the modeled or empirical estimates in Phoenix are more accurate. In particular, dry deposition, which can contribute significantly to N inputs in arid systems, can be difficult to quantify empirically (Lohse et al. 2008). On the other hand, modeled estimates of wet and dry N deposition are often generated from regional- scale models, such as the Community Multi-scale Air Quality (CMAQ) model, or national monitoring networks for wet and dry deposition. Both the CMAQ model and monitoring networks are limited in their spatial resolution or ability to capture spatial heterogeneity, particularly near urban systems and in patchy arid systems, and have been found to over- or underestimate local-scale deposition (Holland et al. 2005; Fenn,
Haeuber, et al. 2003; Fenn et al. 2013; Bettez and Groffman 2013). Overall, the large discrepancy between modeled and empirical rates of N deposition highlights the difficulty in estimating deposition in urban aridland systems (Lohse et al. 2008).
While wet deposition during rainy seasons is expected to be an important seasonal input to arid systems (Báez et al. 2007; Li et al. 2013), dry deposition is thought to contribute up to 80% of atmospheric inputs to arid landscapes where limited precipitation is spatially and temporally heterogeneous (Hanson and Lindberg 1991; Lohse et al. 2008;
Fenn et al. 2010; Li et al. 2013). Dry N deposition, however, is challenging to quantify because of short gaseous atmospheric life spans, volatilization, saturated leaf surfaces, or biological uptake (Hanson and Lindberg 1991; Lovett 1994; Asman, Sutton, and
Schjørring 1998; Golden et al. 2008; Fenn et al. 2013). For example, throughfall monitoring—a method used in many ecosystems to measure wet and dry ammonium
84 (NH4) and nitrate (NO3)—may underestimate N deposition in arid systems when leaf surfaces become saturated with dry deposition during long periods without rain (Fenn et al. 2000; Fenn et al. 2009). Dry deposition can be estimated from the inferential method based on atmospheric concentrations of ammonia (NH3), nitrogen oxides (NOx), and nitric acid (HNO3) and the deposition velocity (Vd) of each N compound. Deposition velocities—requiring site-specific meteorological, vegetation, and landscape characteristics—can also be difficult to quantify and compound the uncertainty of dry deposition estimates for a region (Wesely and Hicks 2000; Krupa and Legge 2000).
More accurate N deposition estimates may be obtained by employing multiple empirical monitoring techniques that can be extrapolated locally and regionally with modeled N deposition estimates (Holland et al. 2005; Fenn et al. 2013). Quantifying both wet and dry N fluxes and capturing their seasonal variability is important for accurately estimating total ecosystem inputs and effectively mitigating and managing the resulting ecological consequences.
Given the largely unknown rates of total N deposition—and dry deposition in particular—in arid systems and the uncertainty in drivers and individual sampling methods, I asked, what are spatial and temporal patterns and dominant drivers of total
(wet and dry) N deposition in arid regions? Using the Phoenix metropolitan region where N deposition estimates are uncertain and the surrounding Sonoran Desert as a case study, I employed multiple sampling approaches to quantify N deposition. I used throughfall (wet and dry) and bulk (wet) collectors to examine long-term seasonal dynamics of N deposition in urban and outlying desert locations. I also employed the inferential method to compare dry deposition across seasons and urban and outlying
85 desert locations. Inferential estimates of dry deposition can provide an indication of the upper bounds of potential dry deposition in the system, which can then be used in combination with other methods to approximate total deposition.
Overall, I expected total N deposition to differ spatially and seasonally, largely dependent upon precipitation patterns and proximity to urban areas. Similar to other arid urban regions, I expected N deposition to exceed the desert critical load. In particular, I expected the highest deposition rates to occur in the urban region near emissions sources.
Following modeled estimates, I also expected deposition to be high in outlying desert locations to the east where downwind deposition was predicted to be higher than in other remote areas. Due to seasonal differences in wet inputs, I expected higher total deposition during the rainy seasons. Yet due to the patchy precipitation, I hypothesized dry deposition would account for a majority of total annual deposition.
To test the drivers of N deposition, I hypothesized that not only the amount of precipitation but also the timing of rain events and location relative to urban areas would be an important predictor of N deposition patterns. For example, I expected metrics of extended dry periods, such as the number of antecedent dry days and the number of consecutive rain-free days, would improve the estimates of total wet and dry (throughfall) deposition, whereas wet (bulk) deposition would be better predicted by total precipitation and the number of rainy days. In addition to rainfall, I expected site-specific meteorological variables (temperature and relative humidity) and urban characteristics
(e.g. traffic density) to be strong drivers of total N deposition patterns.
86 METHODS
Site Characteristics
The Phoenix metropolitan area situated in the northern Sonoran Desert is an ideal case study to examine N deposition in drylands as a result of urbanization. Phoenix has been characterized as one of the fastest growing urban areas in the US (Gober 2006).
With rapid urban sprawl housing a population of over 4 million, the Phoenix metropolitan region and the surrounding Sonoran Desert experience the effects of land use conversion, the urban heat island, and elevated reactive N gas emissions (Brazel et al. 2000; Baker et al. 2001; Grimm and Redman 2004). The ecological effects of N deposition are expected to lead to changes in primary production and community composition with cascading effects on ecosystem services such as ecosystem N retention (Hall et al 2011). Yet, rates of atmospheric N inputs from urban and agricultural emissions to the ecosystem are largely unknown, and accurate estimates are further compounded by the difficulty in quantifying dry deposition. Accurate N flux estimates are important for predicting the ecological consequences of N inputs to ecosystems and evaluating the effectiveness of conservation and N emission control policies (Holland et al. 2005; Lovett 2013).
The Phoenix metropolitan region provides a unique study system for examining the effects of urbanization on arid ecosystems. Municipal ordinances have preserved large remnant patches of native Sonoran Desert within the metropolitan area, making an urban- rural comparison possible between urban remnant desert and outlying desert locations.
The Sonoran Desert is characterized by unique vegetation, precipitation and wind patterns. Drought tolerant cacti (e.g. Saguaro, Carnegiea gigantea), shrubs (e.g. creosote
87 bush, Larrea tridentata), and winter herbaceous vegetation dominate the landscape.
Average annual precipitation is 208 mm with bimodal winter (October – March) and monsoon summer (June – September) seasonal rains (NOAA 2010), and there is a precipitation gradient across the valley with highest rainfall at slightly higher elevations to the east (Table 1). In addition, the average minimum and maximum daily temperatures in the region are 17 and 31oC, respectively (NOAA 2010).
Finally, the Phoenix valley has complex topography resulting in atmospheric mixing patterns that shift daily and seasonally affecting spatial and temporal distribution of atmospheric compounds (Nunnermacker et al. 2004; Wang and Ostoja-Starzewski 2004;
Lee, Fernando, and Grossman-Clarke 2007). The unstable summer atmosphere leads to significant atmospheric mixing throughout the valley, while the more stable winter atmosphere and winter inversion create less vertical mixing. For example, the winter atmosphere tends to settle in the urban core in the early morning until mid-day warming of the eastern mountains and winds from the west causes the atmosphere near the mountains to rise and mix, at which time it is replaced by atmospheric wind from the city center. In the evening, cooling temperatures cause the air to settle back into the valley.
Long-term N deposition monitoring
Monitoring site characteristics: In order to examine my questions regarding long- term N deposition across seasons and urban and remote locations, I continuously monitored N deposition at 15 sites within the 6400 km2 Central Arizona-Phoenix Long- term Ecological Research (CAP LTER) area from March 2006 – December 2013. Five monitoring sites were located within urban desert remnant patches (urban locations, n =
88 5) and 5 sites each were located to the west and east of Phoenix in protected desert preserves (outlying locations, n = 10; Figure 1; Table 1). One urban site (Mountain View
Park, MVP) had limited sampling after 2007 because of frequent human disturbance of field equipment. All sites have similar sandy loam soils, vegetation dominated by drought tolerant creosote bush, and are part of an extensive manipulative N and phosphorus addition experiment measuring ecosystem variables across the landscape
(Hall et al. 2011; Sponseller et al. 2012).
Measuring bulk and throughfall NO3 and NH4 deposition: With ion exchange resin
(IER) collectors, I measured bulk (wet) deposition in interplant open spaces and throughfall (wet and dry) deposition under the dominant shrub. Both bulk and throughfall collectors were deployed with 2 replicates at each site. Collectors were deployed continuously over four 3-month sampling intervals per year (n = 31 sampling intervals from 2006 - 2013) – approximately January to March, March to June, June to
September, and September to December. Collectors were in the field for an average deployment of 89 days (range: 35 – 121 days in the field across all sites and periods).
Replicate bulk and throughfall subsamples were averaged for each site and period (15 sites x 31 periods, n = 465). Dry deposition was estimated by taking the difference of throughfall and bulk deposition (after averaging the subsamples) for each site and period.
Similarly, the ratio of NH4:NO3 in throughfall and bulk deposition was calculated for each site and period after averaging subsamples. The actual number of samples varies for throughfall (n = 419), bulk (n = 410) and dry deposition (n = 395) due to some missing, broken or contaminated field samples, as well as the minimal sampling at MVP.
89 Bulk and throughfall IER collectors were built with a modified design following Fenn and Poth (2004) and Simkin and colleagues (2004). For each sampling period, IER collectors were made with hydrochloric acid-washed supplies. Each IER collector consisted of a 12” PVC pipe filled with 60 mL Monosphere Dowex Resin (Dow
Chemical Company). The resin-filled pipe was closed on one end with a PVC cap with 5
– 7 holes drilled for drainage and filled with glass wool to prevent resin loss. On the other end of the resin-filled PVC pipe, nitex screening was used to prevent debris in the resin and a plastic funnel was placed onto the top. The funnel was covered with mesh and a bird spikes to prevent debris contamination of the resin.
In the field, bulk collectors were installed 1.5 meters above the ground on rebar in open space areas without canopy cover. Throughfall collectors were placed under L. tridentata, where 2-ft holes were dug and filled partly with aquarium rocks to allow for drainage during rain events. All IER collectors (including the funnels and the L. tridentata branches for throughfall collectors) were rinsed with 500 mL di-ionized water before being collected at the end of the sampling period. In addition, field blanks were prepared without a funnel and instead capped on both ends. Field blanks were deployed at one urban (DBG) and two outlying sites (WTM and LDP) with three replicates at each site for both bulk and throughfall field blanks.
After exposed samplers were collected in the field, NH4 and NO3 ions were extracted from each resin sample with 200 mL 2M potassium chloride (KCl) solution. KCl-resin slurries were shaken for one hour and then filtered through Whatman 42 filters pre- leached with KCl. In addition, three KCl extract blanks were prepared. All KCl extracts were analyzed on the continuous flow injection Lachat QuikChem 8000 (Lachat
90 Instruments) for NH4 and NO3. Data were corrected for extract blanks and field blanks from the from their corresponding landscape location. There were no field blanks deployed between March 2006 – December 2007, thus samples from these periods were corrected with an average of field blanks from the following four sampling periods.
-1 -1 With N concentrations (mg N-NO3 L and mg N-NH4 L ) from the KCl extract solutions, I calculated N deposition (kg N ha-1 yr-1) for each site and period. N deposition was estimated based on the volume of extract solution (0.2 L), open area of the funnel
(314 cm2) on each collector, and the number of days each sampler was deployed in the field.
Estimating dry deposition with multiple sampling approaches
Monitoring site characteristics: Using passive filter samplers and the inferential method, I estimated gaseous N concentrations and dry N deposition, respectively, at two locations (open symbols Figure 1). This two-site design was chosen to compare N concentrations and dry deposition at an urban location with expected high atmospheric N concentrations and an outlying desert location with expected low N concentrations. The anticipated difference in N deposition makes an ideal set-up for comparing the dry deposition estimated by IER collectors (described above) and the inferential method. In addition, the outlying location at Lost Dutchman State Park (LDP) is approximately 65 km east of Phoenix and is co-located with one of the long-term monitoring sites described above. The urban site is located in a dense residential neighborhood near central
Phoenix. The urban site is also a monitoring location for the Arizona Air Quality
91 department. Both sites have a meteorological tower operated by the CAP LTER that provided site-specific data for calculating deposition velocities of each N species.
Measuring gaseous NH3, NOx, and HNO3 concentrations and estimating dry deposition: At both locations, I measured concentrations of NOx, NH3, and HNO3 using co-located passive samplers. NOx, NH3, and HNO3 are expected to make up the predominant reduced and oxidized N gases that deposit to surfaces and have ecological consequences (Holland et al. 2005). Passive samplers were deployed for consecutive 2 -
3 week intervals over two summer seasons (3 and 4 summer sampling intervals in 2010 and 2011, respectively) and two winter seasons (4 and 3 sampling intervals in 2010-2011 winter and 2011-2012 winter, respectively). NH3 and NOx samplers were deployed with
2 replicate samplers per site per sampling interval (NH3: 4 replicates/site, NOx and NO2:
2 replicates each/site). Similar to NOx and NH3, four replicate HNO3 samplers were deployed during each sampling interval at a site. Replicates were averaged by site and sampling interval.
Ammonia and NOx (NOx and NO2) concentrations were measured using Ogawa
Teflon passive samplers and Ogawa impregnated filter pads (Koutrakis et al. 1993;
Roadman et al. 2003). Ambient NH3 concentrations were collected on filter pads coated with citric acid, which forms ammonium citrate in the presence of NH3. Ambient NOx concentrations were collected by simultaneously exposing NO2 and NOx filter pads that together were used to calculate total NOx (details of the calculations below). Ambient
NO2 filters were coated with triethanolamine (TEA) and NOx filter pads with TEA and
PTIO (2-phenyl-4,4,5,5-tetramethylimidazoline-3-oxide-1-oxyl). TEA reacts to form
- nitrite (NO2 ) and PTIO is a free radical reagent to scavenge NO and NO2 simultaneously.
92 Nitric acid samplers were designed following methods developed by Bytnerowitz and colleagues (2005) set-up using nylon membrane filters (Pall brand Nylasorb nylon membrane filters, 1.0µm, 47mm) to collect ambient air absorbed as NO3. For each site and sampling period, one additional sampler was set-up as a field blank for each filter type.
Passive samplers were all transported to and from the field in a sealed bag within a sealed plastic container. In the field, passive samplers were installed under a protective cover to block direct sun and rain at approximately 2 meters above ground and away from tall vegetation and structures. Field blanks were prepared and transported to the field sites the same as samplers that were deployed. However, field blanks were returned to the lab immediately after the field samples were deployed and remained sealed on a lab bench at room temperature during the sampling period.
At the end of each sampling period, exposed filters were transported to the lab, where they were transferred to a sealed 20-ml acid-washed glass vial. Dry filters were stored in the freezer until analysis. Each filter was extracted separately with double de-ionized water (DI) directly before analysis and shaken on a shaker table at 165 rpm for 15 minutes. NH3, NO and NO2 filters were extracted with 8 mL DI water and HNO3 filters with 20 mL DI water. NH3 and HNO3 samples were filtered through a 0.02 �m syringe filter (Acrodisc 13mm, 0.2 �m nylon syringe filters to prevent clogging in the analytical instrument) and analyzed in duplicates on a Dionex ion chromatograph (Dionex
Corporation): NH3 samples for ammonium and HNO3 samples for nitrate. NO and NO2 filters were analyzed for nitrite on the continuous flow injection Lachat Quikchem
(Lachat Instruments). The field blanks were extracted and analyzed the same as the
93 exposed filters. I also analyzed extraction blanks (unexposed “blank” filters that were extracted and analyzed with each set of samples). I averaged duplicate samples from the same sampling period and corrected each with the corresponding blank.
Calculating N concentrations and dry deposition: Gaseous N concentrations and dry deposition rates were calculated using a variety of methods. Ammonia concentrations were calculated following Ogawa protocol (equation 1) based on exposure time, extract concentration and volume and the site-specific diffusion coefficient based on
-1 meteorological variables. In equation 1, NH4 (�g mL ) is the concentration of the
-1 -1 sample extract, 17.04 (�g �mol ) is the molecular weight of NH3, 14.01(�g �mol ) is molecular weight of N, and 24.45 is the constant conversion factor for volume to mass of an ideal gas at standard temperature and pressure. The alpha conversion factor converts
NH4 to NH3 concentration by molecular weight and includes the mass transfer diffusion coefficient – or sampling rate – that was calculated based on the geometry of the sampler and the molecular diffusivity of the gas using site-specific meteorological variables during each sampling period (Roadman et al. 2003). I calculated the � conversion factor for NH3 following Roadman and colleagues (2003) based on average temperature during the sampling interval.
!" !!"∗!"# !" !" !"! ∗ !"#$%&# !"#$%& !" ∗ ! !".!" !".!" �� !"! = !" !" ∗ !"#$ ∗ !"#$ (1) ! !! !!"#$%&' !"#$ !"# !! !".!" !" !".!" !"#$ !"#$
Nitrogen dioxide gaseous concentrations were calculated similarly to NH3 following
-1 Ogawa protocol (Equation 2). NO2 (�g mL ) is the concentration of the sample extract
-1 from the NO2 filter pad, 46.01(�g �mol ) is the molecular weight of NO2, and the �
94 conversion factor was calculated specifically for each site and sampling period based on temperature, relative humidity and vapor pressure of water.
!" !!"∗!"# !" !" !"! ∗ !"#$%&# !"#$%& !" ∗! !".!" !".!" �� !"# = !" !" ∗ !"#$ ∗ !"#$ (2) ! !! !"#$%&'( !"#$ (!"#) !! !".!" !" !".!" !"#$ !"#$
Total NOx gaseous concentrations were determined by summing NO and NO2 concentrations. Since there is no � conversion factor for NOx (the combination of NO and NO2), the concentration of NO and NO2 was determined separately and then summed. NO was calculated by subtracting the extract concentrations of NO2 from the corresponding NOx sample from the same site and interval. Then, following Equation 2,
NO concentration (�gN m-3) was calculated using an NO specific � conversion factor and the molecular weight of NO.
HNO3 concentrations were calculated using the Bytnerowitz et al (2005) calibration curve of absorbed NO3 on each filter when exposed to particular HNO3 dose in controlled conditions (slope = 69.498 (hour m-3); equation 3).
!" !!"# !" !" ∗ !"#$%&# !"#$%& !" ∗ !".!"# !".!" !"# ! !" !! !"#$ ���! ! = ∗ !" (3) ! !"#$%&'( !"#$ (!!"#) !".!" !"#$
Using the inferential method, I estimated dry N deposition based on the concentrations from the passive samplers (equations 1 – 3) and estimated deposition velocities (Vd, cm sec-1) for each gaseous N species (equation 4). I used deposition velocities previously estimated by the scholars in the CAP LTER network for the outlying desert site (Gonzalez, unpublished data; Table 2). While deposition velocities vary between sites due to different vegetation structure and meteorological conditions, the deposition velocities I used for each N species were comparable to those used in other
95 arid and urban-arid ecosystem studies (Wesely and Hicks 2000; Zhang, Brook, and Vet
2003; Pan et al. 2012; Delon et al. 2012; Li et al. 2013).
!! !! ���������� �� � ℎ� �� = � ������������� ∗ �! (4)
To examine the accuracy of long-term N deposition estimates from IER collectors within my study site, I compared total deposition estimated from multiple sampling approaches. At the urban and outlying location, I co-located bulk and throughfall IER collectors with the passive filter samples described above. I planted a L. tridentata shrub at the urban comparison site (an empty dirt lot with meteorological monitoring equipment) in order to collect throughfall using the same method as in the outlying location. The IER collectors were deployed continuously over the same sampling intervals each season (installed at the beginning of the first sampling period with passive filter samplers and collected at the end of the last sampling interval per season). The
IER collectors were analyzed for NH4 and NO3 with the same methods described above.
In order to compare total (wet and dry) N deposition estimates, I compared total N deposition estimated from throughfall (wet and dry) to total N deposition estimated by adding wet bulk deposition with dry deposition calculated from the inferential method.
Meteorological and urban site characteristics
Meteorological variables: I gathered several site characteristics, including meteorological variables that were averaged (or summed for precipitation) over each sampling interval. Precipitation, relative humidity, temperature and wind speed and direction were downloaded from the nearest meteorological monitoring stations (Table 1;
FCDMC 2013; see Appendix 1 for meteorological stations). In order to address the
96 temporal variability in precipitation during sampling intervals, I calculated 5 precipitation metrics, in addition to total precipitation, for each sampling interval. Metrics included 1) the number of rainy days per interval, 2) a ratio of the number of rainy days to rain-free days in each sampling interval, 3) the longest number of consecutive rain-free days per interval, 4) the antecedent dry days, which was calculated as the number of consecutive dry days before the first rain event for each sampling interval (including rain-free days from the previous period), and 5) the antecedent dry days specific to the period. The distinction between the last two metrics is expected to be important for explaining variability in throughfall versus bulk deposition, respectively, where bulk deposition is expected to be dependent only upon precipitation while throughfall may capture accumulated dry deposition from the previous sampling interval.
Urbanization variables: I also calculated several anthropogenic related site characteristics. These include housing density, traffic density, and percentage of urban land use within a 10 km buffer area around each site (Table 1). The number of households in each buffer area was calculated with 2010 Census block data including all census blocks that overlapped the perimeter of the 10 km buffer (using the spatial join tool in ArcGIS 10.0). I calculated housing density by summing the total number of households and dividing by the land area (households km-2). Similarly, I calculated traffic density using 2008 average weekday traffic counts, including heavy and light duty traffic on freeways and arterial roads, modeled from the TransCAD travel demand model
(Maricopa Association of Government’s Transportation Division; modeled data shared by Ron Pope, Maricopa Air Quality Department). Using ArcGIS, the traffic density was calculated by summing the traffic count or all the roads within the 10 km buffer around
97 each site (including road segments overlapping the perimeter of the buffer based on
Spatial Join tool) and dividing by the total length of roadways within the buffer.
Data analyses
Statistical analyses were conducted using R (R Core Team, 2014). Dependent variables were transformed to meet basic parametric assumptions to compare N deposition between locations, seasons, and years using analysis of variance. Aggregating temporal data to examine spatial patterns, I used a two-way ANOVA to compare N deposition among sites and regions (outlying west, urban and outlying east). I then aggregated across sites to compare inter and intra annual temporal variability of N deposition.
I used multiple linear regression analyses to determine the main predictors of N deposition (throughfall, bulk, and estimated dry) across the Phoenix metropolitan region.
To test the importance of the timing of precipitation, I first ran multiple linear regression analyses with only the precipitation metrics. Next, I included all site characteristics into the initial model. I used a backward stepwise approach, such that predictor variables were removed from the full model stepwise to determine the most predictive power and parsimonious model based on AIC scores, adjusted R2 and reduced multi-collinearity among independent variables.
For N concentrations and dry N deposition measured between 2010 – 2012, I compared HNO3, NH3, NOx individually and total N (HNO3 + NH3 + NOx) as concentrations and deposition by site location and season using a two-way ANOVA.
98 Throughfall measurements were expected to capture both wet and dry deposition in an ecosystem. Yet, throughfall has the potential to underestimate dry deposition in arid systems where dry deposition can saturate leaf surfaces during extended periods without rain (Fenn et al 2009). In order to examine this potential underestimation in throughfall deposition, I compared two estimates of total (wet and dry) N deposition measured with co-located samplers. Specifically, I compared total throughfall N deposition with total deposition estimated by adding wet deposition collected in bulk IER samplers and dry deposition estimated from the co-located passive gas collectors. Based on these alternative methods of calculating total N deposition, I calculated the percent of total N deposition underestimated by throughfall following equation 5.
% underestimate by throughfall = 100 – (throughfall ÷ (bulk + dry inferential)) (5)
RESULTS
Long-term N deposition
From 2006 – 2013, mean total inorganic N (NH4 + NO3) throughfall deposition in the region was 3.3 (+/- 0.1) kg N ha-1 yr-1 and ranged from 0.2 – 18.0 kg N ha-1 yr-1 (Table 3).
Throughfall deposition, measuring both wet and dry deposited N, was significantly greater than wet bulk deposition (2.0 +/-0.1 kg N ha-1 yr-1, ranging between 0.07 – 11.8 kg N ha-1 yr-1). Estimated using the difference between throughfall and bulk collectors, dry deposition was 42% of total deposition (0-97%; Table 3) and the median dry:wet deposition across all sites of was 0.83 (0 - 29.5).
99 Across the region, NH4 fluxes were greater than NO3. The average NH4:NO3 ratio was 2.8 (+/-0.2) and 1.5 (+/-0.1) in throughfall and bulk deposition, respectively, across all sites (Table 3). On average, NH4 was 68% (36-97%) of total throughfall deposition and 53% (0-86%) of wet bulk deposition. Additionally, variability of NO3 fluxes (65%
CV) was lower than NH4 fluxes (93% CV) in throughfall across sites and years, and was similarly true for wet bulk and dry deposition (Table 3, Appendix 2).
Spatial variability of N deposition: Averaged across seasons and years, total (NH4 +
NO3) throughfall deposition was significantly different among sites (two-way ANOVA, p
= 0.006) and minimally different among regions (p = 0.06). Total urban N throughfall was minimally greater than both outlying west (p = 0.1) and east (p = 0.09). Across sites, total throughfall deposition was significantly greater at MVP than two outlying sites
(outlying west EMW and outlying east SRR, Table 3). However, N deposition collection at MVP ended in 2007 due to human destruction of field samples in the urban park.
When MVP is removed from the analysis, the differences between region are no longer significant (p = 0.15) and throughfall deposition at the DBG was minimally significantly greater than at SRR (p = 0.1).
On the other hand, the patterns of NH4:NO3 and NH4 and NO3 individually in throughfall differed from total (NH4 + NO3) throughfall deposition. The ratio of
NH4:NO3 in total throughfall did not differ significantly among sites (two way ANOVA, p = 0.2). Yet, the outlying west desert region had significantly higher NH4:NO3 than both the urban and outlying east desert regions (p <0.001, Table 3). Examining NH4 and
NO3 deposition individually, average NO3 in throughfall was significantly greater in the city than in outlying sites to east or west of Phoenix (p < 0.001), while NH4 in throughfall
100 did not differ between regions (p = 0.14; data not shown). Across sites, urban DBG and
PWP and one outlying east site (MCN) had greater NO3 throughfall deposition than outlying west EME and EMW (p = 0.004). NH4 was significantly greater at MVP than outlying site EME and SRR (p = 0.02).
Total N (NH4 + NO3) in wet bulk deposition was not significantly different among sites (two-way ANOVA, p = 0.9), but was significantly lower in the outlying west region than urban or outlying east region (p = 0.02, Table 3). NH4:NO3 ratio in wet deposition followed a similar pattern; it did not differ among sites (two-way ANOVA, p = 0.12), but was significantly lower in the outlying west location than outlying east (p = 0.009).
Dry deposition was more variable than throughfall or bulk deposition. Total (NH4 +
NO3) dry N deposition was significantly lower in the outlying east desert than urban or outlying west desert regions (two-way ANOVA, p < 0.001, Table 3). Total dry deposition also varied by site (p < 0.005, Table 3). DBG (urban), SNW and SNE (west sties) were significantly greater than outlying east: LDP, SRR and outlying west: EME.
Temporal variation of N deposition: N deposition varied both intra-annually and inter-annually. Seasonally across all sites, NO3, NH4 and total N (NH4 + NO3) throughfall was greatest between June to September (p<0.001; Figure 2). NH4 and total
N deposition in all other times of the year (January-March, March-June, and September-
December) did not significantly differ. In contrast, NO3 in throughfall was significantly lower between March and June than other periods. Similarly, wet bulk total (NH4 +
NO3) deposition was the greatest during the summer monsoon season of June –
September, while the relatively dry season between March and June had significantly lower deposition than January – March and June – September (p < 0001). Dry total (NH4
101 + NO3) deposition was significantly greater in summer monsoon season between June –
September than all other seasons (p = 0.01)
Significant inter-annual variation was predominately a result of deposition rates measured in 2006, which had an anomalously high period of N deposition between June
– September 2006 (June-September 2006 average throughfall across all sites was 12.7 +/-
0.8 kg N ha-1 yr-1 compared to the same period during other years that ranged between 2.2
- 7.1 kg N ha-1 yr-1; Figure 2). Averaged across sites, total N throughfall was significantly greater in 2006 than all other years (p<0.001). When this period was removed as an outlier, N deposition in 2006 was still significantly greater than in 2009 (p
= 0.04) a year of below-average precipitation (Figure 2). The differences in total N throughfall deposition across all other years were minimal and not significant. There were minimal significant differences in wet bulk deposition across years (p = 0.06), but there were no significant differences among years in post-hoc pairwise tests. However, dry deposition was significantly lower in 2009 than in 2008 (p = 0.04).
Drivers of long-term N deposition: I expected the variability in the timing of the precipitation to be an important driver of N deposition in arid systems; thus, I analyzed throughfall, bulk, and dry N deposition based on precipitation metrics alone (i.e., excluding other site-specific variables). To compare the relative importance of timing in rain events, I compared standardized beta coefficients (st ß) from a best fit multiple regression model. The initial full model started with all 6 precipitation metrics
(Throughfall ~ Precipitation + Rainy days:Dry days + # Rainy Days + Longest consecutive dry days + Antecedent Dry Days + Antecedent dry days by period), and I removed non-significant independent variables stepwise based on AIC scores. For
102 2 throughfall, the best fit model (Table 4, F(5,412) = 60.6, p < 0.001, adj R = 0.42) included all the precipitation metrics, except antecedent dry days by period. Based on the standardized beta coefficients of each variable, the ratio of rainy days to dry days (st ß =
0.82) and number of rainy days per period (st ß = -0.49) had the greatest relative importance and all other variables were less important (total precipitation (st ß = 0.19), # consecutive dry days (st ß = -0.20), and antecedent dry days (st ß = 0.15)). The most parsimonious model predicting bulk deposition was a better fit (i.e. higher adjusted R2) than throughfall and included all precipitation metrics (Table 4, F(6,402) = 89.1, p<0.001, adj R2 = 0.56). Similar to throughfall, the ratio of rainy days to dry days (st ß = 1.1) and number of rainy days per period (st ß = -0.85) were the most important predictors of wet deposition, followed by total precipitation (st ß = 0.31), the antecedent dry days (st ß =
0.21), the longest consecutive number of dry days (st ß = -0.20), and antecedent dry days within the period (st ß = -0.13). Unsurprisingly, the final model predicting dry deposition from precipitation metrics explained little variation and only included two predictors: the number of rainy days per period (st ß = 0.30) and the number of antecedent dry days (st ß
2 = 0.10; Table 4, F(2, 392) = 17.8, p < 0.001, adj R = 0.08)).
In addition to the timing of precipitation, I also expected site-specific factors related to meteorological variables and urbanization to be important predictors of N deposition.
To test this, I used a more extensive multiple regression model following the same backward stepwise procedure as above. The full initial models included the precipitation metrics (as described above), meteorological variables (average temperature, relative humidity, wind speed and wind direction), as well as site characteristics such as elevation, housing density, traffic density, and percent urban land cover within a 10 km
103 buffer area around each site (Table 5). Throughfall was best predicted by a model including independent predictors of temperature (st ß = 0.50), relative humidity (st ß =
0.36), ratio of rain days to dry days (st ß = 0.32), the number of rainy days (st ß = -0.31), total precipitation (st ß = 0.27), as well as the longest consecutive number of dry days, percent urban land cover, average wind speed, and elevation (Table 5; F(9,369) = 52.9, p
<0.001, adj R2 = 0.55).
A mixture of precipitation metrics, meteorological and urbanization factors explained more variability in wet deposition than throughfall or dry deposition models. The most
2 parsimonious model describing wet deposition (Table 5; F(8, 370) = 114, p < 0.001, adj R
= 0.71) including the number of rainy days (st ß = -0.69), the ratio of wet days to dry days (st ß = 0.68), temperature (st ß = 0.55), percent urban land cover (st ß = 0.39), relative humidity (st ß = 0.37), housing density (st ß = -0.34), precipitation (st ß = 0.33), and longest number of consecutive dry days (-0.19). The variability in dry deposition, on the other hand, was not well predicted by the final model (Table 5; F(6, 357) = 12.2, p <
0.001, adj R2 = 0.16). The predictors in the dry deposition model included housing density (st ß = 0.60), traffic density (st ß = -0.33), number of rainy days (st ß = 0.30), average wind speed (st ß = 0.17), temperature (st ß = 0.14), and elevation (st ß = - 0.11).
Gaseous N concentrations and dry deposition
Nitric acid concentrations: Gaseous HNO3 concentrations ranged between 0.1 – 0.7
µg N m-3 in the urban location and 0.1 – 0.5 µg N m-3 in the outlying location. While concentrations were low, in a two-way comparison between location and season, HNO3 was significantly greater in the urban (0.4 +/- 0.1 µg N m-3) than outlying location (0.3
104 +/- 0.04 µg N m-3, p=0.01), and in the summer across both sites (0.5 µg N m-3+/- 0.03) than winter (0.2 µg N m-3 +/- 0.02, p < 0.001; Figure 3). There was not a significant interaction between season and location.
Ammonia concentrations: Average gaseous NH3 concentrations ranged between 4.3 –
11.1 µg N m-3 in urban location and 0.3 – 2.0 µg N m-3 in outlying location. Urban
-3 gaseous NH3 concentration (6.6 µg N m +/-0.6) was significantly greater than outlying concentrations (1.0 µg N m-3 +/- 0.2, p < 0.001, Figure 3). In two-way ANOVA, gaseous
NH3 did not significantly differ between seasons and there were no interactions between season and location (p > 0.1).
Nitrogen oxide concentrations: Urban gaseous NOx concentrations ranged from 2.0 –
16.5 µg N m-3 and were significantly greater (9.1 µg N m-3 +/- 1.4) than outlying east concentrations (0.6 µg N m-3 +/- 0.2, range 0.3 – 1.0 µg N m-3, p < 0.001, Figure 3). As a main effect, season was only minimally significant, with NOx concentration greater in the winter (6.1 µg N m-3 +/- 1.7) than in the summer (3.0 µg N m-3 +/- 1.3, p = 0.07) due to variability between sites and seasons. However, in the urban location NOx concentration in winter was significantly greater than in summer, whereas in the outlying location the opposite trend was significant (p = 0.03).
Dry deposition of gaseous nitrogen species: Dry deposition fluxes were calculated with the inferential method using the empirical concentrations measured at the urban and outlying site described above. I estimated deposition fluxes using the average deposition velocity from a local Sonoran Desert study as the most probable, site-specific deposition velocity for this study (Table 2). Across both sites and years, total dry N deposition was comprised of 41% (+/- 3%) NH3, 36% (+/- 5%) HNO3, and 22% (+/- 2%) NOx. The
105 percent of NH3 deposition was significantly greater in the winter than summer season (p
= 0.003), and in the urban location compared to outlying desert (p < 0.001, Figure 4).
HNO3 followed the opposite pattern: significantly greater in the summer than winter (p <
0.001) and in the outlying compared to the urban location (p < 0.001; Figure 4). NOx, however, was significantly greater in the urban than the outlying desert location (p<
0.001; Figure 4), but did not differ by season. Total urban dry N deposition (HNO3 +
-1 -1 NH3 + NOx, 6.4 +/- 0.4 kgN ha yr ) was significantly greater than total outlying dry N deposition (1.8 +/-0.2, p < 0.001; Figure 5, Table 6). When averaged across locations, differences by season were minimal, but winter tended to have lower rates of deposition than summer, particularly at the outlying site (Figure 5).
Total N deposition comparison
N deposition estimates from co-located throughfall and bulk collectors: During the corresponding sampling periods in the summer and winter of 2010 - 2012, I measured
NH4-NO3 fluxes with ion exchange resin samplers. Similar to long-term N deposition estimates with IER samplers in urban and outlying locations, average throughfall deposition was 3.1 +/- 0.6 kg N ha-1 yr-1. Deposition rates were not significantly different between the urban and outlying sites (p = 0.5), but summer deposition at both sites (urban
= 4.1 +/- 0.6, outlying 4.3 +/- 0.4 kg N ha-1 yr-1) was significantly greater than winter deposition (urban = 2.1 +/- 0.6, outlying 1.2 kg N ha-1 yr-1, p = 0.007, data summarized in
Table 6). Like the long-term bulk N deposition estimates, bulk wet deposition was lower than total throughfall (wet and dry), bulk wet deposition was similar between locations (p
= 0.7), and bulk deposition was significantly greater in the summer than winter season (p
106 = 0.04; Table 6). The throughfall and bulk N deposition estimates described here were also comparable to the throughfall and bulk N deposition estimates from the same periods from the long-term N deposition sampling.
Total N deposition correction: Across sites and years (summer 2010 - winter 2012), total throughfall deposition was significantly less than total deposition estimated by bulk and dry deposition methods (paired t-test, t = 4.4, df = 7, p = 0.003). Overall, throughfall underestimated total deposition by an average of 54%. In other words, N deposition measured in throughfall was 46% of total deposition estimated by adding wet bulk deposition and dry deposition estimated by the inferential method. Throughfall was a better measure of total deposition in the outlying location, where the underestimation of total deposition in the outlying location (41 +/- 21%) was lower than in the urban location
(67 +/- 14%, t = -2.0, p = 0.1). Compared by season, throughfall was a better predictor of total wet and dry deposition in the summer (average 40 +/- 20% underestimation by throughfall) than in the winter (average 67 +/- 14% underestimate by throughfall, t = -2.2, p = 0.07, Table 6). The discrepancy between seasonal estimates may be due, in part, to the high gaseous NOx and NH3 concentrations in the winter (Figures 3 & 5). In addition, the stable atmospheric conditions in the Phoenix region during the winter months suggests the urban atmospheric N is more likely to be deposited locally near the source and lead to greater dry deposition in the city during the winter season.
To better estimate total long-term N deposition patterns, I modeled total N deposition in the region accounting for the underestimation in throughfall. While there were seasonal and regional trends in the underestimation of total deposition from throughfall
(Table 6), the differences between locations or season were not overwhelmingly
107 significant and variances are high. Thus, I used a simple model and predicted new estimates of total N deposition based on the average 54% underestimation across both locations and seasons (predicted N deposition = long-term throughfall / 0.46).
Based on this “correction,” the mean predicted total N deposition in the region increases from 3.3 kg N ha-1 yr-1 (in throughfall) to 7.3 (+/- 0.3) kg N ha-1 yr-1 (range 0.4
– 39.2 kg N ha-1 yr-1). The overall patterns followed those of throughfall. For example, the predicted total N deposition did not significantly differ among regions (urban: 8.2 +/-
0.6, outlying east: 6.7 +/- 0.4, and outlying west: 7.0 +/- 0.5 kg N ha-1 yr-1), but was significantly greater in the summer monsoon season (June – September, 12.7 +/- 0.3 k N ha-1 yr-1, p< 0.001) than in the other seasons (January – March: 5.7 +/- 0.3 N ha-1 yr-1,
March – June: 5.1 +/- 0.3 N ha-1 yr-1, September – December: 5.5 +/- 0.3 N ha-1 yr-1).
DISCUSSION
Patterns of N deposition in a large semi-arid city and surrounding native ecosystem
In the Phoenix metropolitan region and surrounding Sonoran Desert, N deposition rates across the region were surprisingly low. From 2006 - 2013, total wet and dry
(throughfall) N deposition was on average 3.3 kg N ha-1 yr-1 (Table 3). Urban throughfall deposition (3.8 kg kg N ha-1 yr-1) tended to be higher than in the outlying desert (Figure
2, Table 3). Yet, contrary to expectations, the average rates were only minimally different. Similarly, neither total (throughfall) nor wet deposition varied among the 15 sites throughout the metropolitan and surrounding Sonoran Desert (Table 3). While total deposition was comparable between the urban and outlying desert regions, wet deposition
108 was greater in the native desert to the east of Phoenix than the native desert west of
Phoenix (Table 3). Compared to the other sites, the eastern region has a slight elevation gain that creates a precipitation gradient east of the urban region with more frequent rainfall events (Table 1 and 5). In contrast, dry deposition was greater in the outlying region to the west of Phoenix where agriculture is predominant (Table 3).
Observed N deposition rates were consistent with those previously reported in this region using wet-dry buckets from 2000 – 2005. Lohse and colleagues (2008) found
-1 mean wet and dry NH4-NO3 deposition across the region was approximately 4 kg N ha yr-1 with few differences among sites, though dry deposition tended to be highest to the west of the urban center. However, the authors noted N deposition was lower than expected and highlighted the potential underestimation of dry fine particulate and gaseous forms of N deposition by wet-dry buckets in arid systems (Lohse et al. 2008).
Based on deposition in other cities as well as Phoenix specific modeled estimates, I expected significantly higher N inputs than observed, particularly to the east where downwind deposition was expected to be higher. For example, the 2006 CMAQ model estimated average urban deposition in Phoenix to be 18-20 kg N ha-1 yr-1 and previous models estimated similar rates with N deposition in the downwind region as high as 25 kg N ha-1 yr-1 (Baker et al. 2001; Fenn, Haeuber, et al. 2003). The empirical estimates in this study may be lower than modeled deposition because these models incorporate all forms of deposition, including particulate, aerosol and gas phase N that are difficult to quantify in field measurements. Deposition in the Phoenix region was also surprisingly low given the deposition rates reported in other large metropolitan systems, such as Los
Angeles, Baltimore, Boston, and several cities in China. The average deposition rate
109 across these cities was 15 kg N ha-1 yr-1 and often exceeded 30-60 kg N ha-1 yr-1 (Alonso,
Bytnerowicz, and Boarman 2005; L. E. Rao and Allen 2010; Cisneros et al. 2010; Pan et al. 2012; P. Rao et al. 2013; Bettez and Groffman 2013; Li et al. 2013). Higher rates of N deposition in some cities may be explained, in part, by significantly more precipitation, less seasonal variability, a higher proportion of wet deposition, and characteristics related to traffic and housing density. Lower rates of dry deposition in more temperate regions may lead to more accurate total N deposition estimates.
Dry deposition is difficult to estimate in arid and semi-arid ecosystems due to prolonged dry periods, and sporadic and spatially heterogeneous rain during the wetter seasons. I estimated dry deposition with multiple approaches in order to better account for the dry component of deposition. First, I estimated dry deposition from the IER collectors by calculating the difference between throughfall and bulk (wet) measurements, where throughfall is expected to capture wet and dry deposition that accumulates on leaf surfaces above the throughfall collector. Bulk deposition is expected to primarily collect wet deposition. Though some dry deposition may enter the bulk collectors, little dry deposition is expected to collect on the surface of the plastic funnels of bulk deposition collectors. The average rate of dry deposition across the region by the
IER collectors was 1.4 (+/- 0.1) kg N ha-1 yr-1 (Table 3; range 0 – 7.9 N ha-1 yr-1). Mean dry deposition from IER collectors across sites only accounted for 29 – 55% (average 42
+/- 1%) of total deposition (Table 3). While dry deposition estimates from IER collectors were comparable to those reported previously in the Phoenix region (Lohse et al. 2008), they are lower than expected based on estimates up to 80% from other urban arid studies
(Li et al. 2013).
110 Second, I estimated dry deposition based on the inferential method using empirical measurements of gaseous N concentrations in an urban and outlying location. Average dry N deposition estimated by the inferential method (4.4 +/- 0.5 kg N ha-1 yr-1, range 0.8
– 9.3 kg N ha yr) was higher than dry deposition estimated by IER collectors. Dry deposition estimates in the outlying desert were comparable between methods (1.1 and
1.8 kg N ha-1 yr-1 from dry throughfall vs inferential estimates, respectively). Yet, urban dry deposition estimates differed more substantially. Urban dry deposition estimated by
IER collectors was 31% (1.6 kg N ha-1 yr-1) greater than the outlying dry deposition. In contrast, urban dry deposition estimated by inferential method was 72% (6.4 +/- 0.4 kg N ha-1 yr-1) higher than outlying dry deposition estimates from the same method. Overall, with the inferential method, dry deposition was approximately 74% of total deposition in the urban region and 40% of total deposition in the outlying Sonoran Desert.
Patterns of high dry deposition in the city resulted directly from higher gaseous N concentrations in the urban region than the outlying Sonoran Desert (Figure 3). N concentrations in HNO3, NH3, and NOx were all greater in the city than in the outlying
Sonoran Desert (Figure 3). These patterns are similar to those reported in other urban regions, though only two other studies measure all three compounds concurrently (Li et al. 2013; Zbieranowski and Aherne 2012). In particular, NH3 gaseous concentrations were comparable to those previously measured in Phoenix and other arid urban regions
(Watson et al. 1994; Bytnerowicz, Omasa, and Paoletti 2007; Sather et al. 2008; Salem,
Soliman, and El-Haty 2009; Li et al. 2013). Gaseous NOx concentrations were also an important component of the urban atmosphere, particularly during the winter months, and were similar to those collected by the same method in other cites (Sather et al. 2007;
111 Moodley, Singh, and Govender 2011; Li et al. 2013). Higher winter NOx concentrations can be attributed to reduced photo-oxidation of this common pollutant from vehicle exhaust as a result of lower temperatures and fewer daylight hours (Sather et al. 2007;
Afif et al. 2009; Li et al. 2013). While NOx gas concentrations were greater than both gaseous NH3 and HNO3, NOx deposition velocities are low (Table 2) and thus NOx contributes a relatively smaller amount to deposition than gaseous NH3 or HNO3. HNO3 and NH3, on the other hand, are highly soluble, reactive gases that deposit closer to their sources. With high deposition velocities, even relatively small concentrations can significantly affect the total deposition fluxes (Hanson and Lindberg 1991; Zhang, Brook, and Vet 2003; Schwede et al. 2011). Comparable to throughfall NH4:NO3 (Table 3) and as in other studies (Cisneros et al. 2010; Pan et al. 2012), NH3 was the most significant component of dry N deposition in the city, while outlying deposition was primarily made up of HNO3 (Figure 5). In California’s San Joaquin Valley near Fresno, estimates of
-1 -1 summer dry N deposition ranged from 0.4 – 15 kg N ha yr , where NH3 made up between 62-88% of N (Cisneros et al. 2010). In Phoenix, NH3 dry deposition only made up 41% of total dry deposition across seasons (Figure 4). NH3 concentrations and deposition are expected to increase, however, as the projected N emissions in the US are likely to shift from predominantly oxidized N to predominantly reduced N emissions by
2050 (Ellis et al. 2013).
Drivers and sources of N deposition in an arid metropolitan region
Though spatial variation in N deposition was minimal in the Phoenix metropolitan region, N deposition varied seasonally, and seasonal precipitation was an important
112 predictor of deposition. The Sonoran Desert is characterized by bimodal precipitation in the winter and summer. Following rainfall patterns, I expected similar deposition rates in the summer and winter rainy seasons. Yet, the summer monsoon season (June –
September) had higher rates of total (wet and dry) deposition (5.8 kg N ha-1 yr-1) than other times throughout the year, including during the winter rainy season (Figure 2; average range 2.3-2.7 kg N ha-1 yr-1). Like other arid systems, Sonoran Desert summer monsoon precipitation was a strong predictor of summer N deposition (Báez et al. 2007;
Lohse et al. 2008; Li et al. 2013). Unlike other studies that measured wet deposition in rainfall collectors, I examined both wet and total throughfall deposition and found summer precipitation was strongly correlated to both (correlation coefficient 0.68 and
0.63 for summer total (throughfall) and wet (bulk) deposition, respectively, based on
Spearman’s rho rank correlation; Appendix 3). Similar to summer patterns, winter precipitation was also correlated (0.56 correlation coefficient) with winter total
(throughfall) deposition, though winter precipitation was more strongly related (0.83 correlation coefficient) to winter wet (bulk) deposition (Appendix 3). The strength of the winter relationships suggests wet deposition was more dependent on the amount of rainfall during the winter season than total throughfall deposition.
To further investigate the relationship between precipitation and N deposition, I calculated several metrics to capture the timing of rainfall over the integrated 3-month N deposition sampling intervals. In addition to total precipitation, these metrics included the number of rainy days per interval, a proportion of rainy days to dry days in each interval, and the longest consecutive span of dry days per period (Appendix 3). I expected N deposition, and throughfall in particular, to be related to the timing and
113 frequency of precipitation rather than just total precipitation. From best fit multiple regression models, both throughfall and wet bulk deposition were most strongly related to precipitation metrics describing the number of rain events (Table 4; e.g. the proportion of rainy days relative to the dry days and the absolute number of days in which it rained during each sampling interval). Though total precipitation was an important positive predictor of deposition, the number of consecutive dry days was negatively related to deposition indicating less deposition during longer spans without rain. Together, these relationships highlight that aridland total N deposition rates are higher when there are more frequent rainfall events. These trends lend support to the hypothesis that total deposition, in particular the dry component, may be underestimated during long dry periods when leaf surfaces become saturated (prohibiting further collection of N deposition on the surface of leaves during dry periods) or when particulate N volatilizes from leaf surfaces during hot periods.
In addition to precipitation as a primary driver of N deposition, I also examined the influence of other site-specific characteristics, including other meteorological variables, elevation, and factors related to urbanization that were expected to influence N deposition and increase model predictability. Temperature and relative humidity were primary factors explaining throughfall deposition (Table 5). Temperature and relative humidity are seasonally dependent, and along with the precipitation metrics described above, explain the patterns of higher summer N deposition. Total throughfall deposition was also related to the percentage of urban land cover surrounding each monitoring site, wind speed and elevation (Table 5). The same overall set of factors described wet deposition, though the timing of precipitation was relatively more important than temperature and
114 relative humidity in explaining the variability of wet N inputs. On the other hand, long- term dry deposition (calculated as difference in throughfall and bulk deposition) was best predicted by housing and traffic density relative to other site characteristics such as number of rainy days, wind speed, temperature and elevation. The predictors of dry deposition match well with high urban dry deposition estimates based on inferential calculations, where highly reactive NH3 and HNO3 are deposited closer to their sources.
The spatial distribution of NH4 and NO3 and their various gaseous components can be an important indicator of the source of deposition in the region (Holland et al. 2005; P.
Rao et al. 2013; Li et al. 2013). For example, high NH4:NO3 suggests atmospheric NH3 inputs to ecosystems from agricultural sources (N fertilizer applications and animal husbandry), although NH3 can also be a secondary pollutant of motor vehicles. On the other hand, lower NH4:NO3 ratios indicate sources of NOx from industrial combustion that are deposited as NO3 in precipitation. Average NH4:NO3 in long-term throughfall deposition was approximately 2.8 across the Phoenix and outlying desert region, and NH4 was 67% (36-97%) of total throughfall across sites and seasons. While NH4 in thoughfall deposition did not vary among regions, the NO3 deposition was significantly greater in the city than in the outlying desert to the west, indicating fewer sources of combustion in the outlying desert. Similar to the other urban studies, NH4:NO3 in throughfall was lower (2.3) in Phoenix urban sites where housing and traffic density are highest (Table 3,
(Holland et al. 2005; P. Rao et al. 2013; Li et al. 2013; Pan et al. 2012; Bettez and
Groffman 2013; Lovett et al. 2000)). In contrast, the high NH4:NO3 (3.7) west of
Phoenix indicated a likely contribution of NH3 from agriculture in the outlying region more than NO3 from industrial and fossil fuel combustion (Table 3).
115
Underestimation of N deposition in arid systems
The underestimation of N deposition – particularly dry deposition in arid regions – is a common concern (Lohse et al. 2008). To address these uncertainties, I integrated multiple methods to more accurately estimate total and dry deposition. In comparing total deposition estimated by throughfall with total deposition estimated by wet deposition (from bulk measurement) and dry deposition (from inferential estimates), I found that throughfall underestimated total deposition by an average of 54%.
Throughfall was better at estimating total deposition in the outlying desert regions and during the summer months where dry deposition is expected to be lower and more frequent rain events are important for accurately capturing total deposition (e.g. Figure
5). Throughfall has also been reported by other studies to underestimate deposition when compared with alternative wet and dry deposition estimates (Weathers et al. 2000; Fenn et al. 2013). For example, Fenn and colleagues (2013) noted throughfall in western forests may underestimate total wet and dry N deposition by 20-40%, and up to 80% during winter months as a result of N uptake through plant leaves. Throughfall underestimation may result from potential leaf uptake or retention of N by L. tridentata or possible difficulty in rinsing N from sticky L. tridentata leaves. Though L. tridentata leaf uptake and retention were not specifically tested here (see Padgett et al. 1999), leaf uptake was expected to be relatively minor, because throughfall estimates were significantly higher than wet bulk deposition for both NH4 and NO3. Rather, leaf saturation and volatilization from leaf surfaces are likely to be the primary reason for the underestimation of total deposition, especially during long dry periods.
116 To account for the underestimation of long-term N deposition in this arid region, I predicted N deposition estimates based on long-term throughfall measurements and the average percent by which throughfall underestimates total deposition (54%) based on the inferential method. Accounting for the underestimation, average total deposition across the region and study period is predicted to be 7.3 (+/- 0.3) kg N ha-1 yr-1 (range between
0.4 – 39.2 kg N ha-1 yr-1) and average summer deposition is 12.7 (+/- 0.8) kg N ha-1 yr-1
(Figure 6). The corrected deposition estimates more closely approximate those of the
CMAQ model and other arid urban regions, but are still lower than expected for a large metropolitan region. In addition, the predicted long-term N deposition rates across the region are at the upper limit of the expected range (3 – 8 kg N ha-1 yr-1) for aridland N critical loads (Fenn et al. 2010; Pardo et al. 2011). Long-term deposition above the critical load has potentially significant ecological effects on ecosystem processes and primary producer community composition (Brooks 2003; Báez et al. 2007; L. E. Rao,
Allen, and Meixner 2010).
Uncertainties in urban arid N deposition estimates
Despite using multiple methods to estimate N deposition in the region, many uncertainties remain and total N deposition estimates in this study may be conservative for several reasons. First, dissolved organic nitrogen (DON), which in some systems is a significant component of atmospheric N deposition though may be relatively minor in arid regions, was not accounted for in empirical monitoring (e.g. see Jiang et al. 2013;
Cornell 2011; Neff et al. 2002). Second, the variability in atmospheric compounds is also driven by the complex topography and seasonal and diurnal atmospheric mixing
117 patterns in the Phoenix valley (Wang and Ostoja-Starzewski 2004; Nunnermacker et al.
2004; Lee, Fernando, and Grossman-Clarke 2007; Lohse et al. 2008). In the summer, the atmosphere tends to be more unstable causing significant mixing and increased movement and deposition of gaseous particles. On the other hand, vertical flux of gaseous particles is more limited when the atmosphere is more stable during winter months, though the stable atmosphere and winter inversion also allows gaseous concentrations to build up in the city atmosphere. I aimed to account for these factors by including elevation and meteorological variables such as wind speed and direction in analyses. However, the urban topography (i.e. building height) was not accounted for and the average meteorological variables integrated over a 3-month sampling interval likely did not have enough temporal resolution to capture their importance in driving patterns of atmospheric N in the region.
Finally, as noted, dry deposition is difficult to accurately estimate from all methods.
Passive samplers, including throughfall and passive gaseous samplers, have a low temporal resolution such that it is only possible to estimate integrated time-averaged fluxes (Golden et al. 2008). For similar reasons, it is difficult to connect atmospheric concentrations and deposition to vegetation and ecosystem responses (Golden et al.
2008). Overall, this contributes to the uncertainty and underestimation of throughfall rates because of leaf saturation, biological uptake, or volatilization during hot periods.
With the inferential method, there are additional uncertainties associated with deposition velocity estimates, which are highly dependent on the meteorological variables, surface characteristics, and the height and heterogeneity of the surrounding landscape (Wesely and Hicks 2000). Deposition velocities tend to be lower in semi-arid and sparse
118 vegetation areas than in denser canopy forests (Hanson and Lindberg 1991). In contrast, deposition velocities in urban regions tend to be higher (Zhang, Brook, and Vet 2003;
Zhang et al. 2009). Yet, few studies have empirically modeled or tested deposition velocities in arid urban cities, and thus there are potential uncertainties in arid dry deposition estimates.
To calculate dry deposition from the inferential method in this study, I applied an average deposition velocity for each N species estimated for the Sonoran Desert (Table
2). However, even when concentrations are similar, the application of slightly different deposition velocities across locations or seasons can change the deposition estimate by a factor of 2-3 (Schwede et al. 2011). For example, applying the deposition velocities used for estimating deposition in several Chinese cities located in semi-arid regions (HNO3:
-1 1.7, NH3: 0.28, NO2: 0.2 cm sec ; Table 2; Pan et al. 2012; Li et al. 2013), the dry deposition estimates calculated by the inferential method in the Phoenix region would increase by an average of 61% (64% in the city and 58% in desert regions). Based on the higher deposition velocities, the average urban and outlying desert dry deposition could be as high as 10.6 and 2.8 kg N ha-1 yr-1, respectively (compared to 6.4 and 1.8 kg N ha-1 yr-1 estimated with the Sonoran Desert deposition velocities). This further increases the likelihood of significant underestimation of total deposition in throughfall measurements.
For example, underestimation by throughfall is expected to increase to 65% when dry deposition is calculated with the higher deposition velocities compared to 54% underestimation when calculated with Sonoran Desert specific deposition velocities.
Overall, the Sonoran Desert deposition velocities are comparable to those used in the other arid and urban-arid studies and are expected to be the best estimate for this system
119 (Table 2, e.g. Zhang et al 2003, Pan et al 2013, Adon et al 2013, Li et al 2013).
However, estimates with higher dryland deposition velocities highlight the potential upper range of deposition in this large semi-arid metropolitan region.
CONCLUSION
Overall, despite the size and population of Phoenix, N deposition was lower than expected compared to other large arid cities. Lower than expected deposition was likely due, in part, to the difficulty in quantifying dry deposition in arid ecosystems where precipitation patterns are spatially and temporally patchy. Inconsistencies between N deposition sampling approaches reveal the difficulties in accurately estimating dry deposition, as well as how uncertainties related to quantifying site characteristics and deposition velocities can easily confound N deposition estimates.
By accounting for dry deposition with multiple methods, predicted total wet and dry deposition in the region was expected to be in the upper range of the aridland N critical load. Both inferential and throughfall methods indicate the highest deposition rates were restricted to the urban core, though deposition to ecosystems in the outlying region also exceeded predicted critical loads. Despite relatively low levels compared to other arid and urban regions, changes in ecosystem structure and function are likely to occur at the current rates of deposition, particularly in rainy years where microbial communities are expected to be more active (Collins et al. 2008; Hall et al. 2011) and where chronic low levels of N inputs can have significant impacts on community structure (Clark and
120 Tilman 2008). My findings highlight the need for and benefit of mixed methods to more accurately quantify wet and dry N deposition in arid systems.
121 ACKNOWLEDGEMENTS
This work was completed in collaboration with Sharon Hall, Nancy Grimm, Ryan
Sponsellor, and Stevan Earl. Many thanks to Quincy Stewart, Roy Erickson, David
Huber, Sam Norlin, Cathy Kochert, Jennifer Learned, Yevgeniy Marusenko, Brenda
Ramierz, Michelle Schmoker, and Kris Fecteau for help with field work and laboratory analyses. Thank you to Diane Alexander (USFS) for initial guidance on setting up and analyzing the passive samplers, Ron Pope (Maricopa County Air Quality Department) for providing local traffic and air quality data and insightful conversations about Phoenix air quality, and Daniel Henz (Flood Control District Maricopa County) for providing met station data. This work was work supported by the National Science Foundation under grant no. DEB-0423704 and BCS-1026865 Central Arizona-Phoenix Long-term
Ecological Research program and grant no. 0504248, Integrative Graduate Education
Research and Training (IGERT) in Urban Ecology.
122 REFERENCES
Aber, John D., W. McDowell, K. Nadelhoffer, A. Magill, G. Berntson, M. Kamakea, S. McNulty, W. Currie, L. Rustad, and Ivan Fernandez. 1998. “Nitrogen Saturation in Temperate Forests: Hypotheses Revisited.” Bioscience 48 (11): 921–34.
Afif, Charbel, Alain L. Dutot, Corinne Jambert, Maher Abboud, Jocelyne Adjizian- Gérard, Wehbeh Farah, Pascal E. Perros, and Toufic Rizk. 2009. “Statistical Approach for the Characterization of NO2 Concentrations in Beirut.” Air Quality, Atmosphere & Health 2 (2): 57–67. doi:10.1007/s11869-009-0034-2.
Alonso, R., A. Bytnerowicz, and W. I. Boarman. 2005. “Atmospheric Dry Deposition in the Vicinity of the Salton Sea, California—I: Air Pollution and Deposition in a Desert Environment.” Atmospheric Environment 39 (26): 4671–79. doi:10.1016/j.atmosenv.2005.04.016.
Asman, Willem a. H., Mark A. Sutton, and Jan K. Schjørring. 1998. “Ammonia: Emission, Atmospheric Transport and Deposition.” New Phytologist 139 (1): 27– 48. doi:10.1046/j.1469-8137.1998.00180.x.
Báez, S., J. Fargione, D. I. Moore, S. L. Collins, and J. R. Gosz. 2007. “Atmospheric Nitrogen Deposition in the Northern Chihuahuan Desert: Temporal Trends and Potential Consequences.” Journal of Arid Environments 68 (4): 640–51. doi:10.1016/j.jaridenv.2006.06.011.
Baker, Lawrence A., Diane Hope, Ying Xu, Jennifer Edmonds, and Lisa Lauver. 2001. “Nitrogen Balance for the Central Arizona–Phoenix (CAP) Ecosystem.” Ecosystems 4 (6): 582–602. doi:10.1007/s10021-001-0031-2.
Baron, Jill S. 2006. “Hindcasting Nitrogen Deposition To Determine An Ecological Critical Load.” Ecological Applications 16 (2): 433–39. doi:10.1890/1051- 0761(2006)016[0433:HNDTDA]2.0.CO;2.
Bettez, Neil D., and Peter M. Groffman. 2013. “Nitrogen Deposition in and near an Urban Ecosystem.” Environmental Science & Technology 47 (11): 6047–51. doi:10.1021/es400664b.
Bobbink, Roland, K. Hicks, J. Galloway, T. Spranger, R. Alkemade, M. Ashmore, M. Bustamante, S. Cinderby, E. Davidson, and F. Dentener. 2010. “Global Assessment of Nitrogen Deposition Effects on Terrestrial Plant Diversity: A Synthesis.” Ecological Applications 20 (1): 30–59.
Brazel, A., N. Selover, R. Vose, and G. Heisler. 2000. “The Tale of Two Climates - Baltimore and Phoenix Urban LTER Sites.” Climate Research 15 (2): 123–35.
123 doi:10.3354/cr015123.
Brooks, Matthew L. 2003. “Effects of Increased Soil Nitrogen on the Dominance of Alien Annual Plants in the Mojave Desert.” Journal of Applied Ecology 40 (2): 344–53. doi:10.1046/j.1365-2664.2003.00789.x.
Bytnerowicz, Andrzej, Kenji Omasa, and Elena Paoletti. 2007. “Integrated Effects of Air Pollution and Climate Change on Forests: A Northern Hemisphere Perspective.” Environmental Pollution, Air Pollution and Climate Change: A Global Overview of the Effects on Forest Vegetation, 147 (3): 438–45. doi:10.1016/j.envpol.2006.08.028.
Bytnerowicz, Andrzej, Maria Jose Sanz, Michael J. Arbaugh, Pamela E. Padgett, David P. Jones, and Antonio Davila. 2005. “Passive Sampler for Monitoring Ambient Nitric Acid (HNO3) and Nitrous Acid (HNO2) Concentrations.” Atmospheric Environment 39 (14): 2655–60. doi:10.1016/j.atmosenv.2005.01.018.
Cisneros, Ricardo, Andrzej Bytnerowicz, Donald Schweizer, Sharon Zhong, Samuel Traina, and Deborah H. Bennett. 2010. “Ozone, Nitric Acid, and Ammonia Air Pollution Is Unhealthy for People and Ecosystems in Southern Sierra Nevada, California.” Environmental Pollution 158 (10): 3261–71. doi:10.1016/j.envpol.2010.07.025.
Clark, Christopher M., and David Tilman. 2008. “Loss of Plant Species after Chronic Low-Level Nitrogen Deposition to Prairie Grasslands.” Nature 451 (7179): 712– 15. doi:10.1038/nature06503.
Collins, Scott L., Robert L. Sinsabaugh, Chelsea Crenshaw, Laura Green, Andrea Porras- Alfaro, Martina Stursova, and Lydia H. Zeglin. 2008. “Pulse Dynamics and Microbial Processes in Aridland Ecosystems.” Journal of Ecology 96 (3): 413–20.
Delon, C., C. Galy-Lacaux, M. Adon, C. Liousse, D. Serça, B. Diop, and A. Akpo. 2012. “Nitrogen Compounds Emission and Deposition in West African Ecosystems: Comparison between Wet and Dry Savanna.” Biogeosciences 9 (1): 385–402. doi:10.5194/bg-9-385-2012.
Dentener, F., J. Drevet, J. F. Lamarque, I. Bey, B. Eickhout, A. M. Fiore, D. Hauglustaine, et al. 2006. “Nitrogen and Sulfur Deposition on Regional and Global Scales: A Multimodel Evaluation.” Global Biogeochemical Cycles 20 (4): GB4003. doi:10.1029/2005GB002672.
Ellis, R. A., D. J. Jacob, M. P. Sulprizio, L. Zhang, C. D. Holmes, B. A. Schichtel, T. Blett, E. Porter, L. H. Pardo, and J. A. Lynch. 2013. “Present and Future Nitrogen Deposition to National Parks in the United States: Critical Load Exceedances.” Atmos. Chem. Phys. 13 (17): 9083–95. doi:10.5194/acp-13-9083-2013.
124 Fenn, Mark E., E. B. Allen, S. B. Weiss, S. Jovan, L. H. Geiser, G. S. Tonnesen, R. F. Johnson, et al. 2010. “Nitrogen Critical Loads and Management Alternatives for N-Impacted Ecosystems in California.” Journal of Environmental Management 91 (12): 2404–23. doi:10.1016/j.jenvman.2010.07.034.
Fenn, Mark E., JILL S. Baron, EDITH B. Allen, HEATHER M. Rueth, KOREN R. Nydick, LINDA Geiser, WILLIAM D. Bowman, et al. 2003. “Ecological Effects of Nitrogen Deposition in the Western United States.” BioScience 53 (4): 404–20. doi:10.1641/0006-3568(2003)053[0404:EEONDI]2.0.CO;2.
Fenn, Mark E., Richard Haeuber, Gail S. Tonnesen, Jill S. Baron, Susanne Grossman- Clarke, Diane Hope, Daniel A. Jaffe, Scott Copeland, Linda Geiser, and Heather M. Rueth. 2003. “Nitrogen Emissions, Deposition, and Monitoring in the Western United States.” BioScience 53 (4): 391–403.
Fenn, Mark E., and Mark A. Poth. 2004. “Monitoring Nitrogen Deposition in Throughfall Using Ion Exchange Resin Columns.” Journal of Environment Quality 33 (6): 2007. doi:10.2134/jeq2004.2007.
Fenn, Mark E., Christopher S. Ross, Susan L. Schilling, William D. Baccus, Michael A. Larrabee, and Rebecca A. Lofgren. 2013. “Atmospheric Deposition of Nitrogen and Sulfur and Preferential Canopy Consumption of Nitrate in Forests of the Pacific Northwest, USA.” Forest Ecology and Management 302 (August): 240– 53. doi:10.1016/j.foreco.2013.03.042.
Galloway, James N., FRANK J. Dentener, DOUGLAS G. Capone, ELISABETH W. Boyer, ROBERT W. Howarth, SYBIL P. Seitzinger, GREGORY P. Asner, C. C. Cleveland, P. A. Green, and E. A. Holland. 2004. “Nitrogen Cycles: Past, Present, and Future.” Biogeochemistry 70 (2): 153–226.
Golden, Heather E., Elizabeth W. Boyer, Michael G. Brown, Emily M. Elliott, and Don Koo Lee. 2008. “Simple Approaches for Measuring Dry Atmospheric Nitrogen Deposition to Watersheds.” Water Resources Research 44 (4): W00D02. doi:10.1029/2008WR006952.
Grimm, Nancy B., and Charles L. Redman. 2004. “Approaches to the Study of Urban Ecosystems: The Case of Central Arizona—Phoenix.” Urban Ecosystems 7 (3): 199–213. doi:10.1023/B:UECO.0000044036.59953.a1.
Hall, Sharon J., Ryan A. Sponseller, Nancy B. Grimm, David Huber, Jason P. Kaye, Christopher Clark, and Scott L. Collins. 2011. “Ecosystem Response to Nutrient Enrichment across an Urban Airshed in the Sonoran Desert.” Ecological Applications 21 (3): 640–60. doi:10.1890/10-0758.1.
Hanson, Paul J., and Steven E. Lindberg. 1991. “Dry Deposition of Reactive Nitrogen
125 Compounds: A Review of Leaf, Canopy and Non-Foliar Measurements.” Atmospheric Environment. Part A. General Topics 25 (8): 1615–34. doi:10.1016/0960-1686(91)90020-8.
Holland, Elisabeth A., Bobby H. Braswell, James Sulzman, and Jean-Francois Lamarque. 2005. “NITROGEN DEPOSITION ONTO THE UNITED STATES AND WESTERN EUROPE: SYNTHESIS OF OBSERVATIONS AND MODELS.” Ecological Applications 15 (1): 38–57. doi:10.1890/03-5162.
Koutrakis, P., C. Sioutas, S. T. Ferguson, J. M. Wolfson, James D. Mulik, and Robert M. Burton. 1993. “Development and Evaluation of a Glass Honeycomb Denuder/filter Pack System to Collect Atmospheric Gases and Particles.” Environmental Science & Technology 27 (12): 2497–2501. doi:10.1021/es00048a029.
Krupa, S. V, and A. H Legge. 2000. “Passive Sampling of Ambient, Gaseous Air Pollutants: An Assessment from an Ecological Perspective.” Environmental Pollution 107 (1): 31–45. doi:10.1016/S0269-7491(99)00154-2.
Lee, Sang-Mi, H. J. S. Fernando, and S. Grossman-Clarke. 2007. “MM5-SMOKE- CMAQ as a Modeling Tool for 8-H Ozone Regulatory Enforcement: Application to the State of Arizona.” Environmental Modeling & Assessment 12 (1): 63–74. doi:10.1007/s10666-006-9053-7.
Li, Kaihui, Xuejun Liu, Wei Song, Yunhua Chang, Yukun Hu, and Changyan Tian. 2013. “Atmospheric Nitrogen Deposition at Two Sites in an Arid Environment of Central Asia.” PLoS ONE 8 (6): e67018. doi:10.1371/journal.pone.0067018.
Lohse, Kathleen A., Diane Hope, Ryan Sponseller, Jonathan O. Allen, and Nancy B. Grimm. 2008. “Atmospheric Deposition of Carbon and Nutrients across an Arid Metropolitan Area.” Science of The Total Environment 402 (1): 95–105. doi:10.1016/j.scitotenv.2008.04.044.
Lovett, Gary M. 1994. “Atmospheric Deposition of Nutrients and Pollutants in North America: An Ecological Perspective.” Ecological Applications 4 (4): 629–50.
———. 2013. “Critical Issues for Critical Loads.” Proceedings of the National Academy of Sciences 110 (3): 808–9. doi:10.1073/pnas.1219007110.
Lovett, Gary M., Martin M. Traynor, Richard V. Pouyat, Margaret M. Carreiro, Wei- Xing Zhu, and James W. Baxter. 2000. “Atmospheric Deposition to Oak Forests along an Urban-Rural Gradient.” Environmental Science & Technology 34 (20): 4294–4300.
Moodley, Kandasamy G., Shalini Singh, and Suendran Govender. 2011. “Passive
126 Monitoring of Nitrogen Dioxide in Urban Air: A Case Study of Durban Metropolis, South Africa.” Journal of Environmental Management 92 (9): 2145– 50. doi:10.1016/j.jenvman.2011.03.040.
Nunnermacker, L. J., J. Weinstein-Lloyd, L. Kleinman, P. H. Daum, Y. N. Lee, S. R. Springston, P. Klotz, L. Newman, G. Neuroth, and P. Hyde. 2004. “Ground- Based and Aircraft Measurements of Trace Gases in Phoenix, Arizona (1998).” Atmospheric Environment 38 (29): 4941–56. doi:10.1016/j.atmosenv.2004.04.033.
Pan, Y. P., Y. S. Wang, G. Q. Tang, and D. Wu. 2012. “Wet and Dry Deposition of Atmospheric Nitrogen at Ten Sites in Northern China.” Atmos. Chem. Phys. 12 (14): 6515–35. doi:10.5194/acp-12-6515-2012.
Pardo, Linda H., Mark E. Fenn, Christine L. Goodale, Linda H. Geiser, Charles T. Driscoll, Edith B. Allen, Jill S. Baron, et al. 2011. “Effects of Nitrogen Deposition and Empirical Nitrogen Critical Loads for Ecoregions of the United States.” Ecological Applications 21 (8): 3049–82. doi:10.1890/10-2341.1.
Phoenix, Gareth K., W. Kevin Hicks, Steve Cinderby, Johan C. I. Kuylenstierna, William D. Stock, Frank J. Dentener, Ken E. Giller, et al. 2006. “Atmospheric Nitrogen Deposition in World Biodiversity Hotspots: The Need for a Greater Global Perspective in Assessing N Deposition Impacts.” Global Change Biology 12 (3): 470–76. doi:10.1111/j.1365-2486.2006.01104.x.
Rao, Leela E., and Edith B. Allen. 2010. “Combined Effects of Precipitation and Nitrogen Deposition on Native and Invasive Winter Annual Production in California Deserts.” Oecologia 162 (4): 1035–46. doi:10.1007/s00442-009-1516- 5.
Rao, Leela E., Edith B. Allen, and Thomas Meixner. 2010. “Risk-Based Determination of Critical Nitrogen Deposition Loads for Fire Spread in Southern California Deserts.” Ecological Applications 20 (5): 1320–35. doi:10.1890/09-0398.1.
Rao, Leela E., D. R. Parker, A. Bytnerowicz, and E. B. Allen. 2009. “Nitrogen Mineralization across an Atmospheric Nitrogen Deposition Gradient in Southern California Deserts.” Journal of Arid Environments 73 (10): 920–30. doi:10.1016/j.jaridenv.2009.04.007.
Rao, Preeti, Lucy R. Hutyra, Steve M. Raciti, and Adrien C. Finzi. 2013. “Field and Remotely Sensed Measures of Soil and Vegetation Carbon and Nitrogen across an Urbanization Gradient in the Boston Metropolitan Area.” Urban Ecosystems 16 (3): 593–616. doi:10.1007/s11252-013-0291-6.
Roadman, M. J, J. R Scudlark, J. J Meisinger, and W. J Ullman. 2003. “Validation of
127 Ogawa Passive Samplers for the Determination of Gaseous Ammonia Concentrations in Agricultural Settings.” Atmospheric Environment 37 (17): 2317–25. doi:10.1016/S1352-2310(03)00163-8.
Salem, Alaa A., Ahmed A. Soliman, and Ismail A. El-Haty. 2009. “Determination of Nitrogen Dioxide, Sulfur Dioxide, Ozone, and Ammonia in Ambient Air Using the Passive Sampling Method Associated with Ion Chromatographic and Potentiometric Analyses.” Air Quality, Atmosphere & Health 2 (3): 133–45. doi:10.1007/s11869-009-0040-4.
Sather, Mark E., Johnson Mathew, Nghia Nguyen, John Lay, George Golod, Robert Vet, Joseph Cotie, et al. 2008. “Baseline Ambient Gaseous Ammonia Concentrations in the Four Corners Area and Eastern Oklahoma, USA.” Journal of Environmental Monitoring 10 (11): 1319. doi:10.1039/b807984f.
Sather, Mark E., E. Terrence Slonecker, Johnson Mathew, Hunter Daughtrey, and Dennis D. Williams. 2007. “Evaluation of Ogawa Passive Sampling Devices as an Alternative Measurement Method for the Nitrogen Dioxide Annual Standard in El Paso, Texas.” Environmental Monitoring and Assessment 124 (1-3): 211–21. doi:10.1007/s10661-006-9219-4.
Schwede, Donna, Leiming Zhang, Robert Vet, and Gary Lear. 2011. “An Intercomparison of the Deposition Models Used in the CASTNET and CAPMoN Networks.” Atmospheric Environment 45 (6): 1337–46. doi:10.1016/j.atmosenv.2010.11.050.
Simkin, Samuel M., David N. Lewis, Kathleen C. Weathers, Gary M. Lovett, and Kirsten Schwarz. 2004. “Determination of Sulfate, Nitrate, and Chloride in Throughfall Using Ion-Exchange Resins.” Water, Air, and Soil Pollution 153 (1-4): 343–54. doi:10.1023/B:WATE.0000019958.59277.ed.
Sponseller, Ryan A., Sharon J. Hall, David P. Huber, Nancy B. Grimm, Jason P. Kaye, Christopher M. Clark, and Scott L. Collins. 2012. “Variation in Monsoon Precipitation Drives Spatial and Temporal Patterns of Larrea Tridentata Growth in the Sonoran Desert.” Functional Ecology 26 (3): 750–58. doi:10.1111/j.1365- 2435.2012.01979.x.
Vitousek, Peter M., John D. Aber, Robert W. Howarth, Gene E. Likens, Pamela A. Matson, David W. Schindler, William H. Schlesinger, and David G. Tilman. 1997. “HUMAN ALTERATION OF THE GLOBAL NITROGEN CYCLE: SOURCES AND CONSEQUENCES.” Ecological Applications 7 (3): 737–50. doi:10.1890/1051-0761(1997)007[0737:HAOTGN]2.0.CO;2.
Wang, Ge, and Martin Ostoja-Starzewski. 2004. “Influence of Topography on the Phoenix CO2 Dome: A Computational Study.” Atmospheric Science Letters 5 (5):
128 103–7. doi:10.1002/asl.67.
Watson, John G., Judith C. Chow, Fred W. Lurmann, and Stefan P. Musarra. 1994. “Ammonium Nitrate, Nitric Acid, and Ammonia Equilibrium in Wintertime Phoenix, Arizona.” Air & Waste 44 (4): 405–12. doi:10.1080/1073161X.1994.10467262.
Weathers, Kathleen C., G. M. Lovett, G. E. Likens, and R. Lathrop. 2000. “The Effect of Landscape Features on Deposition to Hunter Mountain, Catskill Mountains, New York.” Ecological Applications 10 (2): 528–40.
Weathers, Kathleen C., Samuel M. Simkin, Gary M. Lovett, and Steven E. Lindberg. 2006. “Empirical Modeling of Atmospheric Deposition in Mountainous Landscapes.” Ecological Applications 16 (4): 1590–1607.
Wesely, M. L, and B. B Hicks. 2000. “A Review of the Current Status of Knowledge on Dry Deposition.” Atmospheric Environment 34 (12–14): 2261–82. doi:10.1016/S1352-2310(99)00467-7.
Zbieranowski, Antoni L., and Julian Aherne. 2012. “Ambient Concentrations of Atmospheric Ammonia, Nitrogen Dioxide and Nitric Acid across a Rural–urban– agricultural Transect in Southern Ontario, Canada.” Atmospheric Environment 62 (December): 481–91. doi:10.1016/j.atmosenv.2012.08.040.
Zhang, L., J. R. Brook, and R. Vet. 2003. “A Revised Parameterization for Gaseous Dry Deposition in Air-Quality Models.” Atmospheric Chemistry and Physics 3 (6): 2067–82.
Zhang, L., R. Vet, J. M. O’Brien, C. Mihele, Z. Liang, and A. Wiebe. 2009. “Dry Deposition of Individual Nitrogen Species at Eight Canadian Rural Sites.” Journal of Geophysical Research: Atmospheres 114 (D2): n/a–n/a. doi:10.1029/2008JD010640.
129 region of Phoenix,region AZ (urban) and of in outlying native desert to the east the and west of 1: TABLE #Temperature,)relative)humidty,)precipitation,)wind)speed)and)direction)annual)average)for)2006)Y)2013;)all)meterological)data)from)nearest)FCDMC)met)station)(see)Appendix)1)) ^Traffic)(average)weekday)traffic/mile)of)road))and)housing)density)(average)households/km2))are)based)on)10km)buffer)surrounding)each)site)location.)) in to city calculated was *Distance of center Phoenix, to urban from site location distance Euclidean ArcGISon 9.3 based AZ Outlying)East Urban Outlying)West Location( Table(1: )Site)characteristics)of)remnant)desert)preserves)in)metropolitan)region)of)Phoenix,)AZ)(urban))and)in)outlying)native)desert)to)the)east)and)west)of)the)city.) Site characteristics deposition N monitoring of sites. Site characteristics remnant of desert preserves in metropolitan Usery)Mountain)Park)(UMP) Salt)River)Recreation)area)(SRR) McDowell)Mountain)South)(MCS) McDowell)Mountain)North)(MCN) Lost)Dutchman)Park)(LDP) South)Mountain)West)(SMW) South)Mountain)East)(SME) Piestawa)Peak)(PWP) Mountain)View)Park)(MVP) Desert)Botanical)Garden)(DBG) White)Tanks)Regional)Park)(WTM) Sonoran)National)Monument)West)(SNW) Sonoran)National)Monument)East)(SNE) Estrella)Mountain)West)(EMW) Estrella)Mountain)East)(EME) Site(Name Elevation( (m) 592 434 539 476 620 458 372 456 397 396 454 375 492 382 331 density^( Traffic( 31061 12310 21833 27614 68942 54661 59138 76219 12428 1940 1770 7460 3811 554 616 density^(((( Housing( 204 212 428 658 819 615 35 29 63 49 2 4 0 0 9 Mean(annual( temperature( (oC)# 23.6 21.9 22.8 22.8 21.9 24.7 23.3 23.3 23.3 23.3 23.5 23.9 23.6 24.8 24.8 city. Mean(annual( humidity(
relative( (%)# 30.2 36.8 32.3 32.3 32.1 27.4 29.8 32.5 32.5 36.1 29.9 31.3 32.2 27.8 27.8 Mean(annual( precipitation( (mm)# 205.0 201.5 192.0 237.0 211.4 168.3 144.8 152.4 139.6 154.8 145.3 127.9 119.4 138.8 133.2 Mean(annual( wind(speed( (mph)# 4.3 1.9 5.3 5.3 6.5 2.6 2.3 2.8 2.8 1.2 4.7 4.8 6.7 2.3 2.3 Mean(annual( (degrees)# direction( 180.4 199.4 190.7 190.7 132.7 177.6 191.6 190.2 190.2 206.7 206.5 174.3 191.5 179.8 179.8 wind(
130 velocities (cm sec 2: TABLE Table 2 Table Modeled estimate for PAestimate Pittsburgh, Modeled : Average ecosystems urban and (cm/sec) for semi-arid arid, velocities deposition available) when range (and/or Sonoran Desert, Sonoranthis study Desert, North China, several cities several China, North N DepositionN velocities arid, semi for Location Modeled estimate Modeled estimate Modeled - 1 Toronto, Canada ) from the) from literature arid, semi for Central China Central Central Central Africa Ecosystem TypeEcosystem Semi-arid, urban Semi-arid, urban Semi-arid, Arid & Arid semi-arid Semi-arid Semi-arid Urban Urban Urban - arid and urban ecosystems (0.15 - (0.15 3.0) 0.68 - 0.68 0.99 HNO3 (g) HNO3 - 0.35 1.77 4.7 1.9 1.2 arid, and urban ecosystems. 2 Deposition Velocity / (cm sec) (0.1 (0.1 - 0.19) 0.22 - 0.22 0.32 3.9 3.9 - 19.8 NH3 (g) NH3 0.67 0.28 0.15 0.1 1 . Average (and/or range when available) deposition 0.01 NO
0.15 - 0.15 0.20 0.1 0.1 - 0.18 0.2 0.2 - 1.7 NO2 0.37 0.07 0.14 0.6 0.2 Zhang et al. 2003 Redling et al. 2013 & Zbieranowski 2012 Aherne et Pan al. 2012 Li et al. 2013 et al. 2012; Delon et al. 2013 Adon Zhang et al. 2003 Gonzalez, Unpublished data Citation
131 Mean (+/ (throughfall 3: TABLE (Min - (Min Max) column#indicate#significantly#different#means#among#sites.# Dry NH4:NO3 is not reported; many dry NH4 and NO3 estimates are zero. % dry = % dry deposition of total deposition (throughfall). of total deposition zero. are % dry = NO3 estimates % dry deposition and dry NH4 many is not reported; Dry NH4:NO3 (SE) mean Overall East Outlying Urban WestOutlying Table&3 ha#yr)#and#NH4:NO3#across#all#seasons#and#years.##Mean#(+/)#1SE)#percent#dry#deposition#at#each#site.##Different#letters#within#a# -
SE) per :#Average#(+/)#1SE)#throughfall#(wet#and#dry),#bulk#(wet),#and#dry#(throughfall#)#bulk)#total#(NH4#+#NO3)##N#deposition#(kgN#
Throughfall, –
bulk) total deposition N (NH cent dry deposition at each site.letters Different within a column indicate significantly means different among sites. SMW EMW WTM SNW MCN PWP UMP MCS SRR SME MVP DBG EME SNE LDP bulk, anddeposition dry N across sites. Average (+/ Total&wet&and&dry&(throughfall) (NH4 + NO3) (NH4 3.4 3.4 (0.4)ab 2.9 (0.4)ab 3.8 (0.4)ab 3.0 (0.4)ab 3.1 (0.4)ab 3.2 (0.4)ab 3.8 (0.5)ab 4.3 (0.6)ab 3.3 (0.6)ab 3.9 (0.5)ab 3.7 (0.6)ab 2.7 (0.4)ab 2.3 2.3 (0.3)a 8.4 (3.1)b 2.6 (0.5)a 0.2#)#18.0 3.3#(0.1)############# 3.1 (0.2) 3.8 (0.3) 3.2 (0.2) Total N 4
+ NO NH4:NO3 0.6 0.6 - 36.8 2.9 2.9 (0.6)a 2.3 (0.2)a 2.1 (0.2)a 2.1 (0.2)a 2.3 (0.2)a 2.2 (0.2)a 2.3 (0.2)a 1.9 (0.2)a 5.1 (2.5)a 2.3 (0.2)a 2.7 (0.4)a 4.1 (0.7)a 5.1 (1.5)a 2.7 (0.3)a 4.0 (1.2)a 2.8 2.8 (0.2) 2.3 (0.1) 2.3 (0.1) 3.7 (0.4) 3 , kg N ha , kg N - 1
yr (NH4 + NO3) (NH4 2.2 2.2 (0.4)a 2.0 (0.4)a 2.0 (0.4)a 2.1 (0.4)a 2.2 (0.4)a 1.9 (0.3)a 2.4 (0.4)a 2.2 (0.4)a 2.9 (NA)a 2.1 (0.4)a 1.7 (0.3)a 1.6 (0.3)a 1.5 (0.3)a 1.7 (0.4)a 2.0 (0.4)a 0.1 0.1 - 11.8 - 2.0 2.0 (0.1) 2.1 (0.2) 2.1 (0.2) 1.7 (0.2) 1 Total N ) and NH Wet&(bulk) 4 :NO -
1SE) throughfall (wet and dry), bulk (wet), and dry 3
NH4:NO3 1.4 1.4 (0.2)a 1.7 (0.3)a 1.5 (0.2)a 1.7 (0.2)a 1.7 (0.2)a 1.4 (0.2)a 1.7 (0.2)a 1.3 (0.2)a 1.4 (0.2)a 1.3 (0.2)a 1.3 (0.3)a 1.0 (0.2)a 1.5 (0.2)a 1.5 (0.2)a at each site averaged across all and seasons years. 0.0 0.0 - 6.4 1.5 1.5 (0.1) 1.6 (0.1) 1.5 (0.1) 1.3 (0.1) ND Estimated&dry&(throughall&;&bulk) (NH4 + NO3) (NH4 1.1 1.1 (0.2)abd 1.1 (0.1)abd 1.3 (0.2)abe 1.1 (0.2)abd 1.6 (0.3)abe 2.2 (0.4)cde 1.0 (0.2)abc 1.7 1.7 (0.3)be 0.8 (0.2)ab 1.8 (0.3)be 2.3 (0.4)de 0.9 (0.2)ab 0.5 0.5 (0.1)a 2.3 (0.3)e 1.4 1.4 (0.1) 1.1 (0.1) 1.6 (0.1) 1.6 (0.1) 0.0 0.0 - 7.9 Total N ND 38.0 (4.7)ac 38.0 (3.9)ac 40.5 (4.5)ac 45.5 (4.3)ac 39.7 (4.1)ab 34.3 (4.1)ac 44.5 (2.5)bc 55.3 (4.5)ac 46.6 (4.7)bc 55.7 (4.2)ac 38.6 31.2 (4.7)a 31.2 (4.7)a 29.2 (4.5)c 59.4 (4.7)a 31.7 0.0 0.0 - 96.7 41.9 (1.2) 41.9 (2.0) 37.0 (2.0) 43.4 (2.2) 45.9 % Dry ND
132
throughfall, bulk a 4: TABLE
Model statistics Antecedent dry days by period Antecedent dry days Longest # consecutive dry days # Rainy days : # Dry rainNumber days of Total precipitation depositionN Predicting deposition N precipitation from variables. Multiple regression parameters and significant predictors of
nd dry deposition the from best
F - 1.0 x 10 1.1 x 10 1.8 x 10 (5, (5, 412) - 0.49 0.02 b 2.5
= 60.6, p < 0.001, adj R = 60.6, p < 0.001, adj
- - Throughfall - 3 3 3
St St - - 0.15 0.20 0.82 0.49 0.19 β
d eposition - - 0.07 0.31 0.55 0.77 0 - .05 fit modelfit each. for All predictors listed are significant (p < 0.05). CI
.95 2
for = 0.42 - -
0.23 1.09 0 0.09 0.21 .33 β
- - 1.2 x 10 1. 1. 1. F ( 3 7 7 - 6, 40 0.0 0.34
b 3.2 x 10 x 10 x 10
2) 4
= 89.1 - - - - 3 3 3 3
Bulk Bulk - - - , p < 0.001, adj R , p < 0.001, adj 0.14 0.21 0.20 0.85 0.31 St St 1.1
β deposition
- - 0.12 0.86 0.19 - 0.08 0.30 1.1
CI
.95 2
= 0. for - - 0.05 0.30 0.43 0.10 0.60 1.3 56 β
F 6.5 ( 2, 392 0.20
0.1 x 10 b
)
= 17.9,
- 4
Dry deposition
p < 0.001, adj R p < 0.001, adj St St 0. 0.10 30
β
0.004 CI 0.2
.95 0
for 2
= 0. N 0.40 0.20 β
08
133
< 0.05). signi 5: TABLE
Model statistics densityTraffic densityHousing Elevation Percent urban land cover Wind speed Relative Humidity Temperature Longest # consecut # Rainy days : # Dry rainNumber of Total precipitation ficant throughfall, predictors N bulk and of dry deposition the from best model fit each. for All predictors listed are sign
Predicting deposition N precipitation from and site characteristics variables. Multiple regression parameters and
days
ive dry days
- - F 1 1. 2.7 x 10 1.6
( .0 x 10 9, 369 5 - 0.04 0.01 0.01 0.96 0.0 b
x 10 x 10
) 1
=
- - Throughfall - - 3 3 52.9 4 3
St St p < 0.001, adj R p < 0.001, adj - - - 0.10 0.16 0.12 0.36 0.50 0.17 0.32 0.31 0.27 β
d eposition - - - 0.18 0.08 0.03 0.23 0.40 0 0.05 0. 0.14 .27 58 CI
.95 2
for = 0. - - -
0.24 0.22 0.50 0.60 0.59 0.40 0.02 0.07 0.04 β 55
- - 3.1 x 10 2.7 x 10 1. 1. F ( 7 8 - 8, 370 0.01 0.02 0.0
b 2.0 x 10 x 10
3 )
= - - - - 4 3 3 3 114
Bulk deposition Bulk , p < 0.001, adj R , p < 0.001, adj - - - 0.34 0.39 0.36 0.55 0.19 0.68 0.69 0.33 St St
β
- - - 0.13 0.25 0.47 0.46 0.23 0.62 0.28 0.91
CI
.95 2
= 0.
for - - - 0.65 0.48 0.63 0.89 0.44 0.07 0.11 0.48 71 β
- - 5.5 x 10 4.8 x 10 4.1 x 10 1.2 x 10 3.0 X 10 3.0 X 2.7 x 10 F (6, (6, 357) b
- - - - = 12.2, p < 0.001, adj R = 12.2, p < 0.001, adj - 4 2 3 2 - 6 4
Dry deposition
- - St St 0.60 0.17 0.14 0.30 0.33 0.11
β
- - CI
0.27 0.05 0.05 0.21 0.63 0.23
.95
for 2
= 0.16 - - β 0.93 0.30 0.24 0.40 0.03 0.00 ificant (p
134 season. the percentage that throughfall (wet and dry) deposition underestimates total (wet bulk + dry) deposition by sampling period outlying and urban location between 2010 ha N 6: TABLE - 1
yr
percent differencepercent that throughfall + vs dry) as the percentage bulk (throughfall estimates of total deposition (wet + dry) deposition underestimates total (wet bulk + dry) deposition by sampling period and season. and period by sampling + dry) deposition total (wet bulk underestimates (wet + dry) deposition Table 4: Table dry gas samplers (inferential method) at an urban and outlying location between 2010 - 2012. Calculated - Calculated 2012. 2010 between location outlying and at urban an method) (inferential samplers dry gas - Location 1 ) measured with co Outlying Total deposition N estimates multiple from methods, and percent difference indeposition N estimate Urban N deposition (kg N ha yr) measured with co-located throughfall and bulk IER collectors and passive passive and IER collectors bulk and throughfall co-located with yr) (kg N ha measured N deposition Season Summer Summer Winter Winter -
located throughfall and bulk IER collectors and passive gaseous samplers (inferential method) at an 2010 2010 2010 2010 2011 2011 2011 2011 Year wet and dry) – (throughfall
2012. Percent difference in total deposition N estimates (th Total 1.0 3.1 3.1 5.1 1.2 1.4 3.6 4.7 Deposition Estimates inferential dry) 11.8 (4.0 + 7.8) 7.3 7.3 (1.1 + 6.2) 9.0 (2.6 + 6.5) 7.4 (2.1 + 5.3) 2.5 (1.2 + 1.3) 4.0 (2.9 + 1.1) 4.9 (2.6 + 2.3) 5.9 (3.5 + 2.5) (bulk wet + Total understimate Throughfall % difference in% difference methods 86 65 58 57 54 64 25 21 underestimate roughfall versus bulkroughfall versus + dry) is throughfall Mean Mean (SD) 76 (15) 76 58 (1) 58 (7) 59 (3) 23 s. N deposition N s. and
( kg 135
Outlying West Urban Outlying East Urban Protected Desert
Figure 1: CAP LTER long-term N deposition monitoring sites within protected desert ecosystems in the Phoenix metropolitan region. Open FIGUREsymbols 1 :indicate Long-term the sitesN deposition in which monitoringwe monitored sites. N concentrations Long-term N depositionin monitoring2010 – 2012 sites to within estimate protected dry deposition. native desert in outlying west, urban, and outlying east locations in the CAP LTER study site. Open symbols indicate the sites used for monitoring N concentrations and estimating dry N deposition in 2010 – 2012.
136
FIGURE 2: Long term seasonal average of N throughfall deposition and precipitation. -1 -1 Mean (+/- 1SE) N throughfall (wet and dry) deposition (NH4-NO3, kg N ha yr , bottom) and precipitation (mm, top) during each season in outlying native desert east and west of Phoenix metropolitan region, and within the urban region. Sampling intervals approximately January – March, March – June, June – September, September – December.
137
12 HNO3 12 NH3 12 NOx
) Location, p = 0.01 Location, p < 0.01 Location, p < 0.01
-3 Season, p < 0.01 Season, p = 0.29 Season, p = 0.07 Loc * Sea, p = 0.47 Loc * Sea, p = 0.14 Loc * Sea, p = 0.03
8 8 8 Summer Summer Summer Summer Winter Winter Winter Winter NH3 ug/m3 NH3 NOx ug/m3 NOx HNO3 ug/m3 HNO3 4 4 4 Nconcentration (ugN m
0 0 0 Outlying Urban Outlying Urban Outlying Urban
-3 Figure 3: Mean (+/- SE) HNO3, NH3, and NOx ambient concentrations (ugN m ) at an FIGUREurban and outlying3: Mean desert ambient location HNO between3, NH 20103, and – 2012.NOx concentrati Significanceons values at urban are results and outlying of 2-way ANOVA comparing location (urban, outlying) and season (summer, winter). location between 2010 – 2012. Mean (+/- SE) HNO3, NH3, and NOx ambient concentrations (µg N m-3) at outlying and urban locations between 2010 – 2012. Significance values from 2-way ANOVA comparing location (urban, outlying) and season (summer, winter).
138
100"100 HNO3&
NOx&HNO 80"80 3 NH3& NOx
60"60 NH3 deposition
40"40 %dryN %&of&Total&Dry&N&Deposi:on& 20"20
0"0 Summer Winter Summer Winter Summer&OutlyingWinter& Summer&Urban Winter&
Outlying& Urban& Figure 4: Mean (+/- SE) percent HNO3, NOx, and NH3 of total dry N deposition at an urban and outlying desert location between 2010 – FIGURE 4: Percent HNO3, NH3, and NOx of total dry N deposition. Mean (+/- SE) 2012. Dry deposition was estimated by the inferential method. percent HNO3, NH3, and NOx of total dry N deposition in outlying native desert and an urban location between 2010 – 2012. Dry deposition was estimated by the inferential method.
139
FIGURE 5: Urban and outlying dry N deposition estimated by the inferential method. Dry N deposition (kg N ha-1 yr-1) calculated by the inferential method from measured N concentrations and average deposition velocities estimated for the Sonoran Desert.
140
FIGURE 6: Predicted total N deposition across multiple seasons. Predicted seasonal -1 -1 deposition (NH4-NO3 wet and dry, kg N ha yr , between January – March, March – June, June – September, and September to December) estimated for each region by accounting for the underestimation of measured throughfall. Solid line indicates mean across all sites and seasons (7.3 kg N ha-1 yr-1) and dashed lines indicate lower (3 kg N ha-1 yr-1) and upper (8 kg N ha-1 yr-1) estimates for the critical load in arid ecosystems.
141
APPENDIX 1
Table 1: Flood Control District of Maricopa County (FCDMC) station IDs for the nearest meteorological station to the corresponding N deposition monitoring site. FCDMC meteorological data were downloaded for 2006 – 2013. If more than one site is listed in a column, data were averaged.
PWP(&(MVP,(Data(from(Phoenix(Dam(station(only(contain(dates(from(2009(^(2013 DBG(&(LDP,(the(Central(Arizona(Phoenix(Long^term(Ecological(Research(Site((CAP(LTER)(maintains(a(meteorological(tower(at(these(locations,(and(data(were(used(when(available CITATION,FOR,DATA East Outlying Urban WestOutlying Location Table,1:, Flood(Control(District(of(Maricopa(County((FCDMC)(station(IDs(for(the(nearest(meteorological(station(to(the(corresponding(nitrogen(deposition(monitoring(site(in(this( Usery Mountain Park (UMP) Mountain Usery (SRR) area Recreation Salt River (MCS) South Mountain McDowell (MCN) North Mountain McDowell Park (LDP) Lost Dutchman West Mountain South (SMW) East (SME) Mountain South (PWP) Peak Piestawa ViewMountain Park (MVP) (DBG) Garden Botanical Desert White Tanks Park (WTM) Regional West Monument National Sonoran (SNW) East (SNE) Monument National Sonoran West Mountain Estrella (EMW) East (EME) Mountain Estrella Nitrogen,deposition,monitoring,site, :(Flood(Control(District(of(Maricopa(County(ALERT(System,(http://www.fcd.maricopa.gov/Rainfall/rainfall.aspx study.((FCDMC(meteorological(data(were(downloaded(for(2006(^(2013.((If(more(than(one(site(is(listed(in(a(column,(data(were(average. Wash,(Cloudburst(Wash Cloudburst(Wash,(Hesperus(Wash Usery(Mtn(Park Saguaro(Lake,(Granite(Reef Fountain(Hills(Fire(Dept,(Lost(Dog( Fountain(Hills(Fire(Dept,( Apache(Trail,(CAP(LTER S(Mtn(Fan(HQ S(Mtn(Park Phoenix(Dam,(Tatum(Basin(Inflow,( Phoenix(Dam,(10(St(Wash(basin,( Osborn(@(64th(St,(Salt(River(@( McMicken(Dam,(White(Tanks(Peak,( Gila(Bend(Landfill,(Maricopa( Mobile,(Upper(Waterman Estrella(Fan,(Waterman(@( Estrella(Fan,(Gila(River(@116th,( FCDMC,Station,name(s) Priest,(Papago(Park,(CAP(LTER( Rainbow,(Waterman Mountains,(Upper(Waterman Holly(Acres Dreamy(Draw(Dam Phx(basin(#3,(Dreamy(Draw(Dam White(Tanks(#3,(Ford(Cyn(Wash 4520, 4740, CAP 4740, 4520, LTER 4815, 4825, 4800 5430, 5425, 5415 6830, 6880, 6890 6675, CAP6675, LTER Precipitation,,,,, station,ID 4660, 5975 5995, 5975 6955, 6900 6965, 6900 6845, 6860 6650 4565 6525 6510 4635 Wind,speed, station,ID 6647 6777 5907 5907 6557 6507 4792 4792 4612 5432 6907 6968 6887 6887 station,ID direction, 6657 6787 5917 5917 6567 6517 4787 4787 4622 5442 6917 6967 6897 6897 Wnd, station,ID humidity, Relative, 6651 6781 5951 5951 6561 6511 4796 4796 4616 5436 6911 6971 6891 6891 Temperature, station,ID 6652 6782 5952 5952 6562 6512 4797 4797 4617 5437 6912 6972 6892 6892
142
APPENDIX 2
Table 1: Descriptive statistics for throughfall (wet and dry), bulk (wet), and dry (throughfall – bulk) deposition (kg N ha-1 yr-1) across all years and locations.
Table+1:*Descriptive*statistics*for*throughfall*(wet*+*dry),*bulk*(wet),*and* dry*(throughfall*+*bulk)*deposition*(kg*N*ha*yr)*across*all*years*and* locations* Throughfall (wet + dry), n = 419 Total NH4 NO3 NH4:NO3 Mean 3.3 2.4 1.0 2.8 Median 2.6 1.8 0.8 2.1 SD 2.7 2.2 0.6 3.2 SE 0.1 0.1 0.03 0.2 %+CV 80% 93% 65% 114% Range 0.2 - 18.0 0.09*+*15.2 0.07-3.1 0.6 - 36.8 Bulk (wet), n = 410 Total NH4 NO3 NH4:NO3 Mean 2.0 1.2 0.7 1.5 Median 1.3 0.8 0.6 1.2 SD 2.0 1.6 0.5 1.2 SE 0.1 0.08 0.02 0.06 %+CV 102% 133% 65% 79% Range 0.07+11.8 0*+10.0 0.05 - 2.9 0+6.4 Dry (Throughfall - Bulk), n = 395 Total NH4 NO3 Mean 1.4 1.1 0.3 Median 1.0 0.8 0.2 SD 1.4 1.1 0.4 SE 0.07 0.06 0.02 %+CV 96% 102% 129% Range 0*+*7.9 0*+*7.5 0*+*2.0
143
APPENDIX 3
Table 1: Non-parametric spearman rho correlation R coefficients between precipitation metrics and total N throughfall (wet and dry) and bulk (wet) deposition. Correlations include annual means (2006 – 2013), and winter (January – March) and summer monsoon (June – September) seasons. All correlations were significant (p < 0.05). See+methods+for+a+description+of+each+precipitation+metric Table+1:+NonCparametric+spearman+rho+correlation+R+coefficients+between+precipitation+metrics+and+total+N+ throughfall+(wet+and+dry)+and+bulk+(wet)+deposition.++Correlations+specific+to+annual+means+(2006+C+2013),+and+winter+ (January+C+March)+and+summer+monsoon+(June+C+September)+seasons.+++All+correlations+are+significant+(p+<+0.05). Longest+#+consecutive+dry+days
Number+of+rainy+days
Antecedent+dry+days Rainy+days:dry+days Precipitation+(mm)
Annual C0.53 C0.19 0.55 0.58 0.52
Throughfall-(wet-and-dry)
Winter C0.31 C0.55 0.55 0.53 0.56
Summer- C0.58 C0.40 0.58 0.66 0.68
Annual C0.68 C0.26
0.69 0.74 0.67
Bulk-(wet)
Winter C0.48 C0.72 0.83 0.83 0.8
Summer C0.61 C0.51
0.46 0.59 0.63
144
CHAPTER 4
CHARACTERIZATION OF THE ‘ECOLOGICAL AIRSHED’ OF NEAR-URBAN
PROTECTED ECOSYSTEMS
ABSTRACT
Atmospheric compounds generated by cities are expected to have significant impacts on protected lands in urban and surrounding native ecosystems. Gaseous reactive nitrogen
(N), ozone (O3) and carbon dioxide (CO2) are generated by human activities and individually act as a resource or stressor to ecosystems, but their co-occurring distribution is unknown. Air quality monitoring programs routinely measure O3 and nitrogen oxides
(NOx) in residential areas to meet human health regulations, while other ecologically important compounds such as ground-level CO2 and reactive N are not monitored in either cities or remote protected lands. Using a spatially extensive design, I characterized the “ecological airshed” of native protected ecosystems within and surrounding Phoenix,
Arizona by quantifying the spatial and temporal distribution of biologically relevant atmospheric compounds, including nitric acid (HNO3), ammonia (NH3), NOx, O3, and
CO2. Additionally, I measured N and O3 concentrations along a 1500 meter transect representing a distance-from-city gradient from the exterior edge to the interior of a large open space native preserve. I found that CO2, reactive N and O3 co-occur in both urban and outlying protected areas at elevated levels likely to affect the ecological structure and functioning of ecosystems. Carbon dioxide and N concentrations, as well as N deposition, consistently co-occur at higher levels within urban open space areas 145 compared to the outlying native ecosystem. For example, summer N deposition to urban ecosystems was up to 42% greater than native ecosystems outside of the city. In the outlying locations, N deposition rates were within or above the estimated critical load (3
– 8 kg N ha-1 yr-1 for deserts) at which ecological impacts from elevated N inputs are expected to occur. Mean daily CO2 concentrations were highest within the city (396 ppm), yet urban CO2 patterns were primarily characterized by high diurnal fluctuations
(average 373 – 443ppm) that reflect anthropogenic activities, such as rush hour traffic.
Though summer average O3 concentrations were 11% lower within the city, urban-based real-time monitors show that ecosystems within and outside of the city experience periods of high acute O3 exposure. In native desert locations to the west of the city, all three pollutants occur at elevated levels where CO2 concentrations mimic those in the urban area as a result of extensive agricultural fields. This study is the first to identify the distinct spatial pattern of co-occurring, ecologically important urban pollutants within protected lands. The affects of the urban atmosphere on ecosystems, in combination with other factors related to urbanization such as elevated temperatures, may lead to longer- term consequences for ecosystem structure, functioning and services. My findings highlight the need for air quality monitoring with an expanded repertoire of compounds known to affect ecosystem services and biological processes in both urban and remote native ecosystems.
146
INTRODUCTION
Cities occupy only a small percentage of Earth’s land surface, but urban-generated compounds affect air quality and ecosystems at local, regional and global-scales
(Dentener et al. 2006; George et al. 2007; Gurney et al. 2009). Atmospheric compounds are frequently transported far beyond the urban or political boundaries in which they are produced and regulated (Akimoto 2003; Monks et al. 2009). Despite the distance from human activities, native ecosystems can be exposed to elevated levels of ecologically important urban air pollutants, such as reactive nitrogen (N), ozone (O3), and carbon dioxide (CO2). Decades of research has found these pollutants act individually as a resource or stressor to ecosystems, and more recent research highlights the potential non- additive effects of the pollutants in combination (Long et al. 2004; Karnosky et al. 2007;
Bobbink et al. 2010). Yet, to date, no studies have estimated the co-occurrence of CO2,
N, and O3 within or near urban areas, and thus the distribution and extent to which these common pollutants co-occur is unknown. In this study, I examined the spatial and temporal distribution of CO2, O3, and gaseous reactive N concentrations and deposition in native desert ecosystems within and surrounding the major metropolitan area of Phoenix,
Arizona. Identifying the land area affected by the co-distribution of these atmospheric compounds within and beyond the urban boundary—the “ecological airshed”—is an important step for setting effective management and conservation strategies to protect native ecosystems and the ecosystem services they provide.
Pollutant standards and regulatory monitoring are increasingly common in both developed and developing cities (Molina and Molina 2004; Parrish et al. 2011), yet the 147 spatial distribution of the urban “ecological airshed” is unknown. Air quality has improved in the last several decades in some large cities, such as Los Angeles and New
York City in the United States (Parrish et al. 2011). However, as cities in developing nations become more industrialized and expand in population and land cover worldwide, air pollutants continue to exceed regulatory standards and are of growing public concern
(Molina and Molina 2004; Parrish et al. 2011; Seto, Güneralp, and Hutyra 2012). Yet, air quality regulations primarily focus on human health and do not account for potential ecosystem impacts (Paoletti and Manning 2007; Lovett et al. 2009; Hidy and Pennell
2010). Further, the spatial resolution of air quality monitoring is often restricted to residential or industrial areas to meet human health regulations with a focus on O3, carbon monoxide (CO), particulate matter, and oxidized sulfur and N compounds
(USEPA Clear Air Act). Other compounds, such as surface level CO2, may be monitored in remote regions for background levels, but are less frequently monitored near cities despite their expected elevated concentrations near human activities and ecological significance. Air quality models exist to address these gaps, but their spatial and temporal resolution is often inadequate to examine the variability and magnitude of ecosystem exposure at local scales. Overall, additional monitoring is needed with an expanded repertoire of compounds to monitor the distribution of the “ecological airshed” in order to identify the affected ecosystems and develop air quality regulations that protect both humans and the environment.
Despite general improvements in air quality, elevated CO2, O3, and N generated from human activities are expected to influence native ecosystems worldwide (IPCC 2013).
For example, human-mediated agricultural and combustion activities have significantly 148 increased concentrations of reactive N gas emissions (Galloway et al. 2008).
+ Atmospheric N deposits to ecosystem surfaces as reduced (ammonium (NH4 ), ammonia
(NH3)) and oxidized forms of N (Lovett 1994). Nitrogen deposition has important consequences on ecosystems and plant community composition by alleviating nutrient limitation and stimulating primary production (Aber et al. 1998; Clark and Tilman 2008;
Payne et al. 2013). Similarly, CO2 is a natural product of ecosystem processes (e.g. respiration), but CO2 emissions are greatest in and near cities—creating an expected
“urban CO2 dome”—as a result of energy combustion (Idso, Idso, and Balling Jr. 2001;
Pataki et al. 2006; George et al. 2007; Gurney et al. 2009; Duren and Miller 2012). CO2 is a greenhouse gas affecting climate change, but also a key resource for photosynthesis
(IPCC 2013). Ozone, on the other hand, is a product resulting from photodegradation
(e.g., oxidation) of CO and volatile organic carbon compounds (VOCs) in the presence of
NOx (Finlayson-Pitts and Pitts 2000). Tropospheric O3 has negative health implications for humans, as well as toxic effects on plant physiology leading to reduced growth or early senescence (Karnosky et al. 2007; Ainsworth et al. 2012). In addition to the individual effects of each pollutant, the pollutants in combination can have synergistic or antagonistic affects on ecosystems and primary producers (Payne et al. 2011; Templer
2013; Smith et al. 2014). Yet, no studies have estimated the co-occurrence of CO2, N, and O3 within or near urban areas in order to determine the land area affected or to estimate ecosystem responses to the co-occurring pollutants on a larger scale.
In order to address this gap, I asked, what is the spatial and temporal co-distribution of CO2, N, and O3 concentrations (the “ecological airshed”) in protected desert areas in and surrounding the major metropolitan region of Phoenix, AZ? Arid and semi-arid 149 ecosystems cover over a third of the globe’s land area on which much of the world’s new urban growth is expected to occur (UN 2009, 2011). Further, dryland ecosystems are predicted to be especially sensitive to regional and global anthropogenic changes, such as elevated CO2 and other urban generated pollutants (Melillo et al. 1993; Pardo et al.
2011). Phoenix, Arizona is situated in the northern Sonoran Desert, where over 4 million people reside with significant impacts on air quality. For example, Phoenix experienced 40 O3 non-attainment days in 2012 and 2013 in which 8-hour average concentrations exceeded the US EPA standard (75 ppb). Predicted mean annual N deposition in the metro Phoenix area is expected to be approximately 7 kg N ha-1 y-1 with seasonal averages ranging between 5 – 13 kg N ha-1 yr-1 (Cook et al. In prep). Finally, intense anthropogenic activities create a “CO2 dome” along roads that also varies with distance from the city and land cover (Idso, Idso, and Balling Jr. 1998; Idso, Idso, and
Balling Jr. 2001; Day et al. 2002; Wentz et al. 2002; Koerner and Klopatek 2002). The spatial and temporal co-distribution of these compounds has not been widely investigated within or beyond the urban boundary.
To examine the spatial and temporal distribution of urban air quality, I monitored local and regional patterns of CO2, O3, and reactive gaseous N concentrations in native protected desert ecosystems in the urban and outlying regions of the Phoenix metropolitan area. Based on ambient gaseous N concentrations, I also estimated N deposition to ecosystems in order to better account for the N fluxes to the landscape relative to the critical load. The N critical load is the rate of deposition at which ecological changes occur in ecosystems and is estimated for desert ecosystems between
3-8 kg N ha-1 yr-1 (Fenn et al. 2010; Pardo et al. 2011). In addition, I examined the 150 small-scale variability of N and O3 concentrations along a transect from the exterior to interior of a large protected desert area within the city. I expected the distribution of each compound to vary as a result of a suite of biophysical and anthropogenic variables
(Wentz et al. 2002; Nunnermacker et al. 2004; Wang and Ostoja-Starzewski 2004; Lee,
Fernando, and Grossman-Clarke 2007). For example, I expected CO2 and N to be highest within urban areas closer to human generated sources of emissions. O3 concentrations, in contrast, were expected to be highest outside of the city due to urban
O3 titration in the presence of higher urban NOx. Despite differences in peak concentrations, I expected native ecosystems within and surrounding the city to be exposed to co-occurring CO2, O3, and N at levels that impact ecosystems.
METHODS
Experimental design and site characteristics
I monitored CO2, O3 and N with co-located passive and active atmospheric samplers in sites with varying proximity to the urban center of Phoenix, Arizona. Phoenix is an ideal location to explore the “ecological airshed” of cities. Several large remnant native desert areas have been preserved within the Phoenix municipal area and at the outlying edge of current urban development, and these preserves make a useful urban-rural gradient for examining the effects of urbanization and proximity to the city on the native ecosystem. Following other ecological studies in the Phoenix region (Hall et al. 2011;
Sponseller et al. 2012), I used a gradient of monitoring sites in the preserves within the city (“urban”) and in protected native desert areas at the edge and outside of the urban boundary (“outlying”). 151
Specifically, I measured O3 and gaseous N concentrations with passive samplers at 10 sites: two outlying west of Phoenix (n = 2), four urban (n = 4), and four outlying east of
Phoenix (n = 4, Table 1, Figure 1). At three of these locations (one outlying west, urban and outlying east site), I continuously monitored CO2 concentration with non-dispersive infrared gas analyzers. All sites except one (the calibration site) are located in protected open space desert preserves that range in size and distance from the urban center (Table
1). In addition, the sites fall along a natural elevation and precipitation gradient that increases to the east of Phoenix (Table 1). All sites have a similar vegetation structure, dominated by drought tolerant shrubs, Larrea tridentata and Ambrosia deltoidea, and diverse annual grasses and forbs in the spring. The calibration site (39th Ave and Earll
Drive) is located in a dense residential neighborhood in west-central Phoenix and was chosen to calibrate samplers with continuous air quality monitors operated by the
Maricopa Air Quality Department (AQD) and temperature and relative humidity monitors located on a meteorological tower operated by the Central Arizona–Phoenix
Long-term Ecological Research (CAP LTER) program. Finally, to examine small-scale spatial variability of pollutants, I examined reactive N and O3 concentrations at five equally spaced points along a 1500 meter transect extending from the edge to the interior of the largest protected open space area within the city (South Mountain, Figure 1).
At all locations, I measured concentrations of NOx, NH3, HNO3, and O3 using co- located passive samplers for five consecutive 2 - 3 week intervals over a summer season
(May 17 – August 14, 2013; see Appendix 1 for specific dates). Due to limited supplies
NH3 was only sampled during the first, third and fourth sampling intervals, while NO and
NO2 were sampled during the second and fifth sampling intervals (Appendix 1). HNO3 152 and O3 were both measured continuously throughout the summer. NH3 samples were deployed with 2 duplicate filters per sampler. NOx, NO2, HNO3 and O3 samples only included 1 filter (or one set of two O3 filters that were analyzed together) per site per sampling interval (NH3: 2 duplicate samples/site, NOx, NO2, HNO3, and O3: 1 sample each/site). CO2 was monitored with infrared gas analyzers nearly continuously from June
2013 to March 2014 (See Appendix 1 for specific dates).
O3 and N field collection and analyses
Ammonia, NOx (NO and NO2), and O3 gaseous concentrations were measured using
Ogawa Teflon passive samplers and Ogawa impregnated fiber filter pads (Koutrakis et al.
1993; Roadman et al. 2003). Ambient NH3 concentrations were collected on filter pads coated with citric acid, which forms ammonium citrate in the presence of NH3. NOx concentrations were captured by simultaneously exposing NO2 and NOx filter pads that together were used to calculate total NOx based on site-specific meteorological variables
(details of the calculations described below). Ambient NO2 filters were coated with triethanolamine (TEA) and NOx filter pads with TEA and PTIO (2-phenyl-4,4,5,5-
- tetramethylimidazoline-3-oxide-1-oxyl). TEA reacts to form nitrite (NO2 ) and PTIO is a free radical reagent to scavenge NO and NO2 simultaneously. Finally, O3 concentrations
- - were collected on filters coated with NO2 that is oxidized to nitrate (NO3 ) when exposed to O3. Nitric acid (HNO3) samplers were designed and deployed following methods developed by Bytnerowitz and colleagues (2005) using nylon membrane filters (Pall brand Nylasorb nylon membrane filters, 1.0µm, 47mm) to collect ambient air absorbed as
- NO3 . 153
Passive samplers were all transported to and from the field in a sealed bag within a sealed plastic container. All passive samplers were installed in the field with a protective cover to block direct sun and rain at 2 meters above ground and in open areas away from shrubs, trees, taller vegetation and built structures (Figure 2). For each compound and sampling period, one additional sampler was set-up as a field blank for each filter type.
Field blanks were transported to the field sites in a sealed bag with the samples being deployed, but were returned to the lab immediately after the field samples were installed and remained sealed on a lab bench at room temperature during the sampling interval.
Filters were extracted for chemical products according to methods described in the
Ogawa protocols (ogawausa.com) and by Bytnerowicz and colleagues (2005). At the end of each sampling period, exposed filters were transferred to and sealed in a 20 ml acid- washed glass vial. Dry filters were stored in the freezer until analysis. Each filter was extracted separately with double de-ionized water (DI) immediately before analysis and shaken on a shaker table at 165 rpm for 15 minutes. NH3, NO and NO2 filters were extracted with 8 mL DI water, and O3 and HNO3 filters were extracted together with 5 and 20 mL DI water, respectively. O3 extracts were diluted in 1:5 dilution in DI water by adding 1mL of extract to 4 mL DI water. NH3, O3, and HNO3 extracts were filtered through a 0.2 �m syringe filter (Acrodisc 13mm, 0.2 �m nylon syringe filters to prevent clogging in the analytical instrument) and analyzed in duplicates on a Dionex ion chromatograph (Dionex Corporation): NH3 samples for ammonium (NH4); O3 and HNO3
- samples for NO3 . NOx and NO2 filters were analyzed colorimetrically for nitrite with
Lachat Quikchem continuous flow injection instrument (Lachat Instruments). The field blanks were extracted and analyzed at the same time as exposed filters, as were extract 154 blanks (additional unexposed “blank” filters) with each set of samples. I averaged duplicate samples from the same sampling period and corrected each with the corresponding field blank.
Calculating O3 and N concentrations and dry N deposition
Gaseous N concentrations and dry deposition rates were calculated using a variety of methods. Ammonia concentrations were calculated following Ogawa protocol (equation
1) based on exposure time, extract concentration and volume and the site-specific
-1 diffusion coefficient based on meteorological variables. In equation 1, NH4 (�g mL ) is the concentration of the sample extract, 17.04 (�g �mol-1) is the molecular weight of
-1 NH3, 14.01(�g �mol ) is molecular weight of N, and 24.45 is the constant conversion factor for volume to mass of an ideal gas at standard temperature and pressure. The alpha conversion factor converts NH4 to NH3 concentration by molecular weight and includes the mass transfer diffusion coefficient – or sampling rate – that is calculated based on the geometry of the sampler and the molecular diffusivity of the gas using site-specific meteorological variables during each sampling period (Roadman et al. 2003). I calculated � conversion factor for NH3 based on Roadman and colleagues (2003) which is based on average temperature during the sampling period.
!" !!"∗!"# !" !" !"! ∗ !"#$%&# !"#$%& !" ∗ ! !".!" !".!" �� !"! = !" !" ∗ !"#$ ∗ !"#$ (1) ! !! !"#$%&'( !"#$ !"# !! !".!" !" !".!" !"#$ !"#$
Nitrogen dioxide gaseous concentrations were calculated similarly to NH3 following
-1 Ogawa protocol (Equation 2). NO2 (�g mL ) is the concentration of the sample extract
155
-1 from the NO2 filter pad, 46.01(�g �mol ) is the molecular weight of NO2, and the � conversion factor is calculated specifically for each site and sampling period based on temperature, relative humidity and vapor pressure of water.
!" !!"∗!"# !" !" !"! ∗ !"#$%&# !"#$%& !" ∗! !".!" !".!" �� !"# = !" !" ∗ !"#$ ∗ !"#$ (2) ! !! !"#$%&'( !"#$ (!"#) !! !".!" !" !".!" !"#$ !"#$
Total NOx gaseous concentrations were determined by summing NO and NO2 concentrations. Since there is no � conversion factor for NOx (the combination of NO and NO2), the concentration of NO and NO2 was determined separately and then summed. NO was calculated by subtracting the extract concentrations of NO2 from the corresponding NOx sample from the same site and period. Then, following Equation 2,
NO concentration (�gN m-3) was calculated using an NO specific � conversion factor and the molecular weight of NO.
HNO3 concentrations were calculated using the Bytnerowitz and colleagues (2005) calibration curve of absorbed NO3 on each filter when exposed to particular HNO3 dose in controlled conditions (slope = 69.498 (hour m-3); equation 3).
!" !!"# !" !" ∗ !"#$%&# !"#$%& !" ∗ !".!"# !".!" !"# ! !" !! !"#$ ���! ! = ∗ !" (3) ! !"#$%&'( !"#$ (!!"#) !".!" !"#$
Ozone concentrations were calculated using Equation 4 from the Ogawa O3 protocol.
In equation 4, NO3 is the sample extract concentration from the O3 filter pads and 18.09 is the constant conversion factor that incorporates the sampling rate (21.8 mL min-1) and molecular weight conversion from NO3 to O3.
!" !" !"# !"! ∗ !"#$%&# !"#$!" !" ∗ !".!" ∗ !""" � ��� = !" ! !" !"! (4) ! !"#$%&'( !"#$ (!"#)
156
As this study is focused on estimating the ecological airshed to determine the extent and impact of urban pollutants on ecosystems, I estimated N deposition rates from N concentrations to use as an indicator of the amount of N reaching the ecosystem.
Nitrogen deposition rates, rather than concentration values, are also useful for determining where N inputs to the ecosystem exceed the critical load data. Using the inferential method, I estimated dry N deposition fluxes (Equation 5) based on the concentrations from the passive samplers (equations 1 - 3) and estimated deposition
-1 velocities (Vd, cm sec ) for each gaseous N species.
!! !! ���������� �� � ℎ� �� = � ������������� ∗ �! (5)
Across all sites and periods, I uniformly applied an average deposition velocity (NO:
-1 -1 -1 -1 0.01 cm sec , NO2: 0.14 cm sec , HNO3: 1.2 cm sec , and NH3: 0.15 cm sec ) previously estimated for each compound at an outlying Sonoran Desert site in the CAP
LTER region (Gonzalez, unpublished data). While deposition velocities likely vary between sites due to differences in vegetation structure and meteorological conditions, the deposition velocities I used for each N species were comparable to those used in other arid and urban-arid ecosystem studies (Wesely and Hicks 2000; Zhang, Brook, and Vet
2003; Pan et al. 2012; Delon et al. 2012; Li et al. 2013).
Monitoring CO2 concentrations
Real-time CO2 concentrations were monitored with 6 non-dispersive infrared gas analyzers (IRGA, CM-0018, CO2meter, Ormand Beach, Fl). These particular CO2 sensors were chosen because they are inexpensive and portable with built-in data loggers for CO2 concentrations, temperature and relative humidity. Recording data every 10 157 minutes, the IRGAs can log for about six-week intervals. Finally, the IRGAs were battery operated for easy deployment in remote desert sites without electricity.
CO2 monitoring occurred at 3 sites in the Phoenix region: one outlying west (EME), one urban (DBG), and one outlying east (MCS; Table 1, Figure 2). At each site, 2 CO2
IRGAs were installed on rebar: one each at 2 and 0.5 meters above the ground in open areas away from shrub and tree vegetation. I monitored CO2 at 2 meters in order to capture ambient concentrations at a similar height to those monitored in other studies
(Idso, Idso, and Balling Jr. 2001) and at 0.5 meters to test if near surface ambient CO2 concentrations —closer to the height of desert herbaceous vegetation—differ from 2- meter concentrations. Sensors were placed under a hard plastic protective shield
(approximately 100 cm diameter and 10 cm side height) to protect it from direct sunlight and rain. CO2 monitoring began on 11 June 2013, and concentrations (ppm) were recorded every 10 minutes continuously until 14 March 2014. Data were downloaded and batteries replaced in each sensor every 2 – 5 weeks throughout the sampling period.
At each site, there were some sensor or data logging errors, creating some gaps in the continuous long-term data (see Appendix 1 for specific dates).
The CO2 IRGAs have a repeatability of +/- 20 ppm and an accuracy of +/- 30ppm
(CO2 meter.com). In order to account for variability among sensors and ensure consistency in readings across sites, I calculated a sensor specific correction (i.e. an offset factor) to apply to the data for each sampling period. Correction factors were determined by simultaneously logging ambient CO2 concentrations with the field CO2 IRGAs and a calibrated LiCOR IRGA (Li-8100 calibrated at LiCOR, Lincoln Nebraska). The Li-8100 is an automated CO2 soil flux system, but for this purpose was set up to draw ambient air 158 and simultaneously record CO2 concentrations with the field sensors every 15 seconds.
A correction factor was calculated based on the average difference in CO2 readings between the Li-8100 and each sampler and applied to the corresponding sampling periods. Correction factors tended to increase over time, but were relatively consistent within a sampler. Correction values ranged among samplers from minimal offset value of -0.8 ppm (average +/- 5.5 SD across sampling periods at EME) to the maximum offset of -33.0 ppm (average +/- 11.9 SD across sampling periods at MCS). A negative offset indicates the field sensors were consistently higher than actual concentrations. The data reported here are corrected CO2 values.
Data analyses
Each pollutant (HNO3, NH3, NOx, total reactive N, O3, CO2) was transformed when needed to meet basic assumptions of normality and homogeneity of variances. When data could not meet these assumptions, I used non-parametric Kruskal Wallis test to compare ranked means. For O3 and N, I first examined differences in mean concentration (O3, HNO3, NH3, NOx, and total N) among sampling locations with one- way ANOVA (O3) and Kruskal Wallis (all N concentrations). Next, I compared summer
O3 concentrations and N deposition among location (outlying west, urban and outlying east) and the sampling period (see Appendix 1) with a two-way ANOVA. I used bonferroni adjustment in the post-hoc comparisons in order to account for multiple pair- wise comparisons. Statistical analyses were conducted using R (R Core Team, 2014).
Ambient CO2 concentrations, monitored over a longer continuous time frame, were analyzed by sampling height (0.5 and 2 meter), site location (EME, DBG, MCS), season 159
(summer: June – August, winter: December – March), diurnal patterns, and monthly concentration. At each site and sampling height, raw 10 minute data were summarized as hourly means to examine the diurnal fluxes (n = 24). I also compared daily means over the entire study period (at 2 meter and 0.5 meter height respectively, n = 263 and 221 at
MCS, n = 195 and 80 at DBG, and n = 247 and 248 at EME, see Appendix 1 for specific dates), and over the summer (June – August 2013) and winter (December 2013- March
2014) seasons, separately. Differences in mean site and monthly concentrations were compared within each season with a two-way ANOVA. I used bonferroni adjustment in the pairwise post-hoc comparisons to account for inflated error due to multiple comparisons among means.
RESULTS
Ozone
During summer (mid-May to mid-August), average two-week summer O3 concentrations ranged from a low of 31.9 ppb at the urban Desert Botanical Garden
(DBG) site to 65.4 ppb at McDowell Mountain South (MCS) to the east of Phoenix
th (Table 2). Mean O3 concentrations were significantly higher at MCS than the urban 39
Ave site, but did not significantly differ among other locations (one-way ANOVA, F(9, 38)
= 2.2, p = 0.04, Table 2). Averaged across sites, summer O3 concentrations were higher in the middle of summer (Period 3, F(4, 45) = 23.4, p < 0.001; Figure 3) and 11% higher on average to the east of the city (48.1 ± 1.4 ppb) than in urban (42.8 ± 1.3 ppb) locations
(two-way ANOVA, F(2, 45) = 9.1, p < 0.001, Figure 4). 160
Nitric Acid
-3 Summer HNO3 concentrations ranged from 0.3 – 1.5 µg N m across sites, but were not significantly different among sites (Kruskal-Wallis, χ2 = 10.1, df = 9, p = 0.3, Table
2). Summer HNO3 two-week average deposition followed the same pattern. Summer
-1 -1 HNO3 deposition ranged from 1.1 kg N ha yr at White Tank Mountains (WTM) to 5.5
-1 -1 kg N ha yr at DBG (Table 3). Averaged across sites, summer HNO3 deposition differed significantly by location relative to the city (two way ANOVA, F(2, 33) = 16.5, p<0.001; Figure 4) and period of summer sampling (two way ANOVA, F(4, 33), = 22.0, p
< 0.001). Summer season HNO3 deposition was significantly greater in the city (3.2 ±
0.3 kg N ha-1 yr-1) than in sites outlying to the east (2.3 ± 0.2 kg N ha-1 yr-1) or west (2.1 ±
0.3 kg N ha-1 yr-1) and was higher in the middle of the summer (periods 2 – 4) than the final period during early August (not shown).
Ammonia
-3 Summer NH3 concentrations ranged from 2.4 – 14.1 µg N m (Table 2). While there was a large range in NH3 concentrations, mean NH3 did not significantly differ among
2 sites (Kruskal-Wallis, χ = 16.5, df = 9, p = 0.06, Table 2). Summer NH3 deposition rates were lowest in the outlying sites to the east and west of the city with deposition as low as
1.2 kg N ha-1 yr-1 (Table 3). The highest ammonia fluxes were 6.7 kg N ha-1 yr-1 at urban
th 39 Ave site (Table 3). Averaged across sites, NH3 fluxes were significantly higher during the third and fourth periods than the first period in early summer (two-way
161
ANOVA, F(2, 20) = 26.2, p <0.001, data not shown). In addition, mean fluxes were significantly higher in the urban (4.2± 0.4 kg N ha-1 yr-1) than outlying desert to the east
-1 -1 (2.1± 0.2 kg N ha yr ; two-way ANOVA, F(2,20) = 26.8, p < 0.001, Figure 4). In contrast, urban NH3 deposition did not differ from the outlying west sites (3.0 ± 0.5 kg N ha-1 yr-1).
Nitrogen oxide
-3 Summer NOx concentrations ranged between 0.1 – 1.2 µg N m with the highest
th concentrations within the urban region (Table 2). NOx concentrations at urban 39 Ave
-3 -3 (1.7 ± 0.2 µg N m ) and DBG (1.1 ± 0.1 µg N m ) were significantly higher than NOx concentrations to the east or west of Phoenix (0.2 – 0.3 µg N m-3; Kruskal-Wallis, χ2 =
16.9, df = 9, p = 0.04, Table 2). When calculated as deposition, summer NOx fluxes were low compared to deposition of other N compounds in this study, ranging from 0.04 in the outlying regions to 0.7 kg N ha-1 yr-1 in an urban site (Table 3). Despite overall low values, when averaged across sites, NOx deposition was greatest in urban compared to outlying locations (two way ANOVA, F(2, 13) = 11.2, p < 0.001, Figure 4). Nitrogen oxide fluxes did not differ by summer period (two-way ANOVA, F(3, 13) = 1.5, p = 0.3).
Total dry N deposition
Total dry N concentration was determined by summing the concentrations of HNO3,
NH3, and NOx for each site and sampling period. Because NH3 and NOx were only sampled during a subset of the season’s sampling periods, the seasonal average of each
162 compound at the corresponding site was substituted for missing sampling periods in order to estimate the total concentration for all sites and periods. Total summer N concentrations ranged between 3.1 – 16.6 µg N m-3 and differed significantly by site
(Kruskal-Wallis, χ2 = 34.3, df = 9, p < 0.001, Table 2). Nitrogen concentrations at 39th
Ave (13.2 +/- 2.8 µg N m-3) were significantly higher than all other sites except urban
DBG (10.8 +/- 1.9 µg N m-3) and SMW (10.3 +/- 3.2 µg N m-3, Table 2). In addition,
DBG and SMW were also significantly higher than three outlying east sites (Table 2).
Total dry N fluxes varied from 2.4 – 11.1 kg N ha-1 yr-1and differed significantly by location relative to the city (Figures 3 and 4). Averaged across sites during the summer, urban total dry N deposition (7.8 ± 0.4 kg N ha-1 yr-1) was significantly greater than to the
-1 -1 -1 -1 east (4.5 ± 0.2 kg N ha yr ) or west (5.1 ± 0.5 kg N ha yr , two-way ANOVA, F(2, 45)
= 72, p< 0.001, Figure 4). Note, these values only account for dry deposition.
At the urban locations, NH3 constitutes an average 53 +/- 6% of total dry N deposition in the region, with the remainder from HNO3 (42 +/- 8%) and NOx (5 +/- 3%).
In the outlying west sites, NH3 also makes up the highest proportion of fluxes (58 +/- 5% of total) compared to 40% (+/- 5) from HNO3. On the other hand, the outlying east sites had more even distribution than the west sites of NH3 and HNO3, 50% (+/- 7) and 48%
(+/- 7) of the total dry deposition, respectively. Among all locations, the proportion of
NOx fluxes were variable but tended to be higher in the city, composing 5% (+/- 3) of total urban N deposition, compared to 1% (+/- 0.2) and 2% (+/- 0.5) in the locations west and east of the city, respectively.
163
Carbon dioxide concentrations
The urban and outlying locations had distinct CO2 patterns that were most pronounced at the 2 meter compared to 0.5 meter sampling height (Figures 5 - 8). At 2 meters above ground, annual mean CO2 concentrations were greater in the urban (396 ppm +/- 29 ppm) and outlying west sites (392 +/- 26 ppm) than CO2 at the outlying east site 382 ppm (+/- 10 ppm). Urban CO2 patterns had greater diurnal variation than concentrations in the outlying locations. In the urban locations, the difference between mean daily minimum (374 ppm) and maximum (443 ppm) values was 69 ppm. In outlying desert to the west (EME), the mean daily minimum and maximum was 377 and
426 ppm, respectively, with a range of 49 ppm. In contrast, the mean diurnal variation in the outlying location to the east (MCS) was only 19 ppm with daily average minimum and maximum ranging between 373 - 393 ppm.
At 0.5 m sampling height, CO2 patterns were more similar among sites (Figures 6 and
8). Averaged across the year, the average concentration was 386 ppm at all sites. Mean daily minimum and maximum values were also similar among sites. The mean daily low concentrations were 372, 373, and 374 ppm at urban, west, and east sites respectively.
The mean daily maximum CO2 concentration was 410 ppm at both urban and west sites and 399 ppm at the east site.
At all sites, CO2 concentrations varied seasonally and tended to be higher in the winter than summer. Average daily winter concentrations were significantly greater at west (404 +/- 34 ppm) and urban (416 +/- 32 ppm) locations than east site (386 +/- 8 ppm; p < 0.001, Figure 4). Winter monthly patterns were not consistent among sites.
Similar patterns among the sites were observed in the summer, yet the daily mean 164 concentration at each site was lower in the summer than winter season (west 385 +/- 11 ppm, urban 387 +/- 15ppm, and east 381 +/- 11 ppm, p < 0.001, Figure 6).
During the summer dates corresponding to N and O3 sampling, mean 2 meter CO2 concentrations were significantly greater in the city (392 +/- 12 ppm) and to the west of the city (387 +/- 11 ppm) than to the east of the city (382 +/- 12 ppm, Figure 4). In addition, diurnal variation in median daily min and max CO2 concentrations was greater at the urban site (71 ppm, range 374 - 445 ppm) compared to the west site (50 ppm, range
371-421 ppm) and east site (45 ppm, range 365 – 410 ppm).
Edge-interior open space transect
Ozone concentrations and N deposition along the edge-interior open space transect were comparable to those in other urban locations in this study. Contrary to expectations, there were few differences among sampling locations on the transect. Ozone concentrations ranged between 40 – 54 ppb during summer 2013, but were not significantly different among sites on the 1500m transect (Table 4, Kruskal-Wallis, χ2 =
1.2, df = 4, p = 0.9). However, similar to O3 trends at the regional-scale, O3 concentrations were highest during mid summer sampling interval (3: 11 June – 12 July) and significantly decreased in the following two sampling intervals (Table 4; Kruskal-
Wallis, χ2 = 12.5, df = 2, p = 0.002).
Total (HNO3 + NOx + NH3) N deposition along the transect ranged between 6.0 –
10.3 kg N ha-1 yr-1, but did not differ significantly among locations (Table 4; Kruskal-
Wallis, χ2 = 1.9, df = 2, p = 0.8). The same pattern held true for each individual species
165
(HNO3, NH3, NOx) of total N deposition. Finally, averaged across all transect sites, there were no significant differences (p > 0.05) in N deposition between summer sampling intervals (Table 4).
DISCUSSION
Spatial and temporal co-occurrence of CO2, O3 and N in an urban “ecological airshed”
I examined the extent and spatial-temporal variability of human-generated pollutants in native ecosystems within and beyond the urban boundaries of the major metropolitan region of Phoenix, Arizona. No previous studies have examined CO2, O3, and reactive N in combination to determine the extent to which they form an “ecological airshed” of biologically important pollutants near and far from their sources.
I found that O3, reactive N, and CO2 co-occur in both urban and outlying protected areas at elevated levels likely to affect the ecological structure and functioning of ecosystems. Carbon dioxide and N concentrations, as well as N deposition, consistently co-occur at higher levels within urban open space areas compared to the outlying native ecosystem. For example, N deposition rates across all locations were within or above the estimated critical load range, but summer N deposition to urban ecosystems was up to
42% greater than native ecosystems outside of the city. Similarly, mean daily CO2 concentrations were highest within the city (396 ppm mean annual CO2). The main difference between urban and outlying CO2 concentrations, however, is reflected in the diurnal patterns influenced by anthropogenic activities. Urban CO2 fluctuates following anthropogenic pulses, such as vehicular emissions from morning and evening rush hour 166 traffic. Despite the distance from major roadways, urban open space preserves experience average daily maximum concentrations of 443 ppm compared to maximum
393 ppm daily maximum values in an outlying native ecosystem. The average daily CO2 fluctuation in cities is 69 ppm, which is 72% greater diurnal variability than in the outlying native ecosystem to the east of the city. Though summer average O3 concentrations were 11% lower within the city, urban-based real-time monitors show that ecosystems within and outside of the city experience periods of high acute O3 exposure.
In addition, long-term O3 exposure at concentrations comparable to the urban region can have significant impacts on ecosystems (Paoletti and Manning 2007).
Despite the distance from the city and human activities, outlying native ecosystems are also within the urban ecological airshed and experience elevated combinations of all three pollutants. Though lower than in the city, N inputs into the outlying native ecosystem occur at rates that have predicted consequences for primary producers resulting in increased annual herbaceous plant growth, a loss of native desert vegetation, and increased fire frequency (Brooks 2003; Báez et al. 2007; Rao, Allen, and Meixner
2010; Vourlitis 2012). Nitrogen deposition in outlying areas co-occurs with elevated summer O3 concentrations to the east of Phoenix, following the dominant meteorological patterns in the valley (Ellis et al. 2000; Lee et al. 2008). In native ecosystems to the west of the city, CO2 concentrations mimic those in the urban area and thus all three pollutants occur at elevated levels. While cities only cover a small percentage of the Earth’s land surface, these results highlight the impact of concentrated urban human activity on ecosystems beyond the urban boundaries.
167
Summer regional O3 patterns
Summer ozone concentrations varied spatially across the urban and outlying native ecosystem. Similar to patterns previously reported in this region, the highest mean O3 concentrations were primarily found beyond the urban borders (Atkinson-Palombo,
Miller, and Balling Jr. 2006; Blanchard, Tanenbaum, and Lawson 2008). In particular, mean O3 concentrations in the native desert to the east of the city were 11% greater than within the city (Figure 4). While vehicle emissions are the main source of anthropogenically produced precursors to O3, the high O3 concentrations outside of the city are likely due to transport of urban O3 plumes to downwind locations and the titration of O3 in the city center. O3 titration leads to a diurnal fluctuation in which urban
O3 concentration dips significantly at night in the presence of high concentrations of NOx
(NO + O3 à NO2 + O2; Finlayson-Pitts and Pitts 2000). With lower traffic density and vehicle emissions in the outlying locations (Table 1), O3 titration is also expected to be lower in the outlying regions.
The passive sampling methods employed in this study have many benefits (e.g. inexpensive, low maintenance) and can be used to estimate ecosystem exposure to pollutants over longer, integrated periods. However, the passive samplers do not capture the diurnal variability, peak concentrations, or acute short-term O3 exposure that can be important for ecosystem consequences. For example, on shorter temporal scales (1 and 8 hour averages), Phoenix urban O3 concentrations were frequently higher than in outlying locations, despite higher long-term averages in the outlying desert regions (Pope and Wu
2014). Similarly, I compared hourly O3 concentrations collected by the Maricopa
County Air Quality Department (AQD) at the calibration site during summer 2013. 168
Hourly O3 concentrations ranged between 67 – 94 ppb and the daily maximum 8-hour O3 concentrations exceeded the 75 ppb EPA standard on 3 days during the study period. In comparison, the mean concentrations across the five summer sampling intervals measured by the Maricopa County AQD (31.2 – 40.1 ppb) were similar to the co-located passive samplers at the same site in this study (32.4 – 41.4 ppb).
Overall, both long-term and short-term O3 concentrations and exposure estimates are important for estimating ecological impacts of O3 (Paoletti and Manning 2007). For example, plants exhibit sensitivity to long-term O3 exposure near 40 ppb, as well as to short-term acute exposure at peak concentrations (Paoletti and Manning 2007; Ainsworth et al. 2012). Further, this highlights that while ecosystem exposure to O3 may be lower in the urban regions compared to outlying regions, the ecological consequences within the city may still be significant. In addition to phytotoxic effects on vegetation, short-term acute exposure that occurs throughout the urban and outlying regions can have negative human health implications.
Regional N concentration and deposition
Summer total N concentrations and deposition rates were higher within the urban open space preserves compared to the outlying native ecosystem. Urban open space areas, which are characterized by higher surrounding traffic and housing density than the outlying locations, had 49% and 33% higher total N (HNO3 + NH3 + NOx) concentrations than the outlying east and west native ecosystems respectively (up to 16.6 µg N m-3;
-3 Table 3). Across all location, NH3 concentrations (ranging between 2.6 – 14.1 µg N m ) were comparable to other urban arid systems and NH3 was the predominant N species 169 contributing over 80% to total N concentrations in the region (Watson et al. 1994;
Alonso, Bytnerowicz, and Boarman 2005; Cisneros et al. 2010; Li et al. 2013). While
NH3 emissions typically result from agricultural sources, NH3 is also a secondary pollutant of motor vehicles (Holland et al. 2005). In contrast, HNO3 and NOx made up a proportionally smaller amount, typically less than 10%, of the total atmospheric N concentrations (Table 3). Though NOx acts as a precursor to O3 formation during warm summer months, summer NOx concentrations were highest in the city near emission sources (Table 3).
Using the inferential method to estimate N inputs to the ecosystem, summer mean dry deposition in the city (7.8 kg N ha-1 yr-1) was greater than in the outlying protected native ecosystem (4.5 – 5.1 kg N ha-1 yr-1). Urban deposition was comparable to dry N deposition estimated using the same methods for previous summers in the region (2010 –
2011 summers 6.4 +/- 0.4 kgN ha-1 yr-1; Cook et al, In prep). However, dry deposition in the outlying region was substantially higher than reported for previous summers (1.8 +/-
0.2 kg N ha-1 yr-1), which is attributed to the increased sampling and spatial coverage in this study.
Despite the variability in N deposition among locations, the total N deposition in both urban and outlying native ecosystems in the Phoenix region was within or exceeded the critical loads. Recent estimates of N critical loads for desert ecosystems range from 2–20 kg N ha-1 y-1, depending on species composition and water availability (Allen et al. 2006;
Fenn et al. 2010; Rao et al. 2010; Pardo et al. 2011). Some models predict a smaller range between 3–8 kg N ha-1 y-1, though the critical loads in deserts have only been estimated by a few studies and uncertainties in the best estimate are high (Fenn et al. 170
2010; Rao et al. 2010). Nevertheless, dry deposition alone in the Phoenix region was estimated to be within this critical load range with potential ecological consequences for the native ecosystem. In addition, wet N deposition can also be an important input, particularly during the monsoon summer season (Báez et al. 2007; Lohse et al. 2008; Li et al. 2013). Previous estimates of summer wet:dry N deposition using co-located passive and bulk ion exchange samplers found dry deposition in the summer months was about 69% of the total deposition in the urban region and 44% in the outlying regions
(Cook et al, In Prep). Based on these ratios and dry deposition measurements from this study, I predicted total wet and dry deposition for the region to be 11.2 (+/- 2.9) kg N ha-1 y-1 in the urban open preserves, with the highest deposition occurring at the most urbanized locations (39th Ave and DBG at 15.1 and 16.1 kg N ha-1 y-1, respectively). In the outlying Sonoran Desert east and west of the city, the average total wet and dry deposition was predicted to be 10.4 (+/- 2.4) and 11.9 (+/- 3.7) kg N ha-1 y-1, respectively.
Thus, accounting for both wet and dry N inputs, the current rates of deposition within the urban region and in outlying protected native ecosystems are likely to affect primary production, species composition, and belowground processes, particularly in years of above average precipitation.
Regional and temporal CO2 concentrations
I examined patterns of ambient CO2 concentrations at 2 and 0.5 meters above ground in remote desert sites and in a desert preserves within the urban ecosystem. With consecutive measurements that capture diurnal, seasonal, and annual variability between native ecosystems in urban and outlying locations, I found the urban CO2 dome varies by 171 season, sampling height, and extent into the native desert. These findings expand our understanding about urban CO2 characteristics beyond those along the major roadways on which it has previously been examined in the Phoenix region.
Urban CO2 concentrations measured at 2 meters above ground were characterized by high diurnal variability (Figure 5, 6). Mean annual urban CO2 concentrations in the city were 396 ppm (+/- 29). Urban CO2 concentrations had 30 and 73% larger daily fluctuations (69 ppm) than the outlying native ecosystem to the west and east of the city, respectively (49 ppm at outlying west site and 19 ppm at outlying east site). While this pattern is true for other urban-rural comparisons (Idso, Idso, and Balling Jr. 2001; Day et al. 2002; Grimmond et al. 2002; Nasrallah et al. 2003; Coutts, Beringer, and Tapper
2007; Helfter et al. 2011; García, Sánchez, and Pérez 2012; Song and Wang 2012), the amplitude of the urban diurnal CO2 flux was greater than reported elsewhere and likely due in part to the complex topography in the Phoenix region (Wang and Ostoja-
Starzewski 2004). Like more remote locations, ambient urban CO2 concentrations follow biological processes of respiration (high pre-dawn concentrations), photosynthesis (low mid-day CO2 concentrations), and the meteorological effects of convective mixing.
However, urban CO2 variability is also influenced by anthropogenic activities, such that urbanized areas with high population, traffic and employment density are important sources of CO2 (Koerner and Klopatek 2002; Wentz et al. 2002; Pataki et al. 2006; Pataki et al. 2007). In particular, vehicular emissions are expected to contribute up to 80% of
CO2 emissions in the Phoenix region (Koerner and Klopatek 2002). Even away from major roadways within the open space preserve, the urban diurnal CO2 fluctuations
172 appear to be augmented by anthropogenic pulses such as morning and evening rush hour traffic (Koerner and Klopatek 2002; Pataki et al. 2007).
Contrary to expectations, ambient CO2 patterns in remote outlying native ecosystems were not always lower than in the city. Characterized by relatively low surrounding housing and traffic density in comparison to the urban location (Table 1), I expected both outlying sites to be on the outskirts of the urban CO2 dome and have overall lower CO2 concentrations and diurnal variability. While mean annual CO2 was lowest in the outlying site to the east of the city (382 +/-10 ppm, 2 meters), CO2 concentrations in the outlying west site (392 +/- 26 ppm) were more similar to urban patterns (Figure 5, 6).
Despite the remote western location, agriculture is a predominant land use in the region west of Phoenix, and agricultural soils can be a significant contributor to CO2 emissions
(Koerner and Klopatek 2002). For example, compared to desert soil CO2 efflux (< 2 g
-2 -1 CO2 m day ), agricultural soils have high CO2 emissions similar to mesic landscapes
-2 -1 (20 - 30 g CO2 m day , Koerner and Klopatek 2002). The agricultural emissions likely contribute to the higher than expected CO2 concentrations at the outlying western location. In addition, the timing of seasonal agricultural activities (e.g. tilling, planting, and irrigation) may affect the diurnal fluctuations of CO2 monitored in the outlying desert. The variation among sites may also be attributed, in part, to the local vegetation structure within each site. For example, while all sites are characterized by a mix of A. deltoidea and L tridentata, the urban site is also situated near a large mesic park, which can be a significant source of CO2 from mesic soils (Koerner and Klopatek 2002). While site differences in vegetation may affect the local scale strength of CO2 sources and sinks,
173 these differences are expected to be minimal in comparison to the human dominated CO2 sources of vehicle and agricultural soil emissions.
Seasonally, winter CO2 concentrations and diurnal variability at all locations were greater than in the summer, which can be explained by a combination of biological and human dominated factors. For example, reduced water stress and mild temperatures during the winter season make biological activity from microbial communities and primary producers an important contributor of ecosystem CO2 fluxes (Noy-Meir 1973).
For example, Pataki and colleagues (2007) estimated up to 60% of CO2 fluxes in Salt
Lake City, UT originated from biological respiration during the growing season in the region. In other large cities, such as London, the elevated winter CO2 concentrations are attributed to increased human consumption of fossil fuels for heating (Helfter et al.
2011). Though less likely in this system, elevated winter CO2 concentrations may result from the winter inversion layer that reduces the atmospheric mixing in the region (Pataki et al. 2007). The stable atmosphere with low wind speeds leads to less vertical mixing and less dispersion of the CO2. Thus, winter CO2 emissions that are generated from human or biological activity “build up,” particularly during evening and early morning hours, such as at the urban and outlying west site (Day et al. 2002). The outlying east location did not follow the same seasonal patterns. This may be a result of the distance from anthropogenic CO2 sources or the higher elevation where the site is less affected by the winter inversion and experiences higher wind turbulence (Table 1).
At 0.5 meters above ground, CO2 concentrations were more consistent; mean annual
CO2 was 386 (+/- 14ppm) at all three locations. The consistent and lower concentrations at 0.5 meter compared to 2 meter concentrations reflect the similarity in local ecosystem 174 characteristics of native desert soils and their low CO2 efflux (Koerner and Klopatek
2002). In addition, these results suggest relatively little atmospheric mixing between the ground level and above shrub canopy (~2 meters), particularly in comparison to the regional atmospheric mixing patterns. The 0.5 meter monitoring better represents the exposure of herbaceous plants and soil microbial communities in native ecosystems.
Thus, despite close proximity to anthropogenic CO2 sources, low lying herbaceous vegetation and the soil microbial communities in urban open spaces may be “shielded” from the elevated CO2 concentrations typically reported in urban regions. While CO2 concentrations are continuing to rise globally (IPCC 2014), these findings have important implications for considering ecosystem responses to future predicted elevated CO2.
Urban “ecological airshed” extends to interior of urban open space preserve
While regional urban and outlying locations differed in atmospheric O3 and reactive
N, variability was minimal on a smaller-scale from the exterior to interior of an open space preserved native ecosystem within the city. I expected the more remote interior of the park to be less influenced by the ecological airshed than the exterior areas. For example, I expected N deposition to decline toward the interior of the open space and away from major anthropogenic sources as a result of high deposition velocities.
However, while total N and O3 concentrations were comparable to those in other urban open space areas, they did not differ along the edge to interior transect. Overall, these findings suggest a well-mixed atmosphere widely distributes urban pollutants. Thus, the urban ecological airshed extends into the interior of even the largest municipal open space preserves of the native ecosystems. 175
CONCLUSION
Global urban land area is expected to triple by 2030 and more than 65% of the world’s population is predicted to live in cities by 2050 (United Nations 2012; Seto,
Güneralp, and Hutyra 2012). Human activities concentrated in growing urban centers are significant sources of atmospheric pollutants, including CO2, reactive N, and O3, that affect air quality, human health, ecosystem services, and ecological functioning (IPCC
2013). My findings highlight that the urban “ecological airshed” extends well beyond the urban boundary where biologically relevant pollutants co-occur in remote areas at levels that affect ecosystem structure and function. It is particularly important to identify the extent of land area exposed to human-generated pollutants in an urban “ecological airshed” as existing urban air quality conditions have been found to affect growth and physiological functioning of primary producers with likely long-term feedbacks on native ecosystem structure, function, and the provision of ecosystem services (Gregg, Jones, and
Dawson 2003; Gregg, Jones, and Dawson 2006). These findings highlight the need and urgency to adopt air quality monitoring and regulations that include a full suite of ecologically relevant compounds to protect the surrounding natural environments and ecosystem services.
176
ACKNOWLEDGEMENTS
This research was completed in collaboration with Sharon Hall. We especially thank
Diane Alexander (USFS) for initial guidance with the passive samplers, Ron Pope
(Maricopa County Air Quality Department) for providing local traffic and air quality data and insightful conversations about Phoenix air quality, and Daniel Henz (Flood Control
District Maricopa County) for providing met station data. Thank you to Stevan Earl,
David Huber, Jennifer Learned, Julea Shaw, Miguel Morgan and Caitlin Holmes for help with methods, field and laboratory work. This work was work supported by the National
Science Foundation under grant no. DEB-0423704 and BCS-1026865 Central Arizona-
Phoenix Long-term Ecological Research program and grant no. 0504248, Integrative
Graduate Education Research and Training (IGERT) in Urban Ecology.
177
REFERENCES
Aber, John D., W. McDowell, K. Nadelhoffer, A. Magill, G. Berntson, M. Kamakea, S. McNulty, W. Currie, L. Rustad, and Ivan Fernandez. 1998. “Nitrogen Saturation in Temperate Forests: Hypotheses Revisited.” Bioscience 48 (11): 921–34.
Ainsworth, Elizabeth A., Craig R. Yendrek, Stephen Sitch, William J. Collins, and Lisa D. Emberson. 2012. “The Effects of Tropospheric Ozone on Net Primary Productivity and Implications for Climate Change*.” Annual Review of Plant Biology 63 (1): 637–61. doi:10.1146/annurev-arplant-042110-103829.
Akimoto, Hajime. 2003. “Global Air Quality and Pollution.” Science 302 (5651): 1716– 19. doi:10.1126/science.1092666.
Alonso, R., A. Bytnerowicz, and W. I. Boarman. 2005. “Atmospheric Dry Deposition in the Vicinity of the Salton Sea, California—I: Air Pollution and Deposition in a Desert Environment.” Atmospheric Environment 39 (26): 4671–79. doi:10.1016/j.atmosenv.2005.04.016.
Atkinson-Palombo, Carol M., James A. Miller, and Robert C. Balling Jr. 2006. “Quantifying the Ozone ‘weekend Effect’ at Various Locations in Phoenix, Arizona.” Atmospheric Environment 40 (39): 7644–58. doi:10.1016/j.atmosenv.2006.05.023.
Báez, S., J. Fargione, D. I. Moore, S. L. Collins, and J. R. Gosz. 2007. “Atmospheric Nitrogen Deposition in the Northern Chihuahuan Desert: Temporal Trends and Potential Consequences.” Journal of Arid Environments 68 (4): 640–51. doi:10.1016/j.jaridenv.2006.06.011.
Blanchard, Charles L., Shelley Tanenbaum, and Douglas R. Lawson. 2008. “Differences between Weekday and Weekend Air Pollutant Levels in Atlanta; Baltimore; Chicago; Dallas–Fort Worth; Denver; Houston; New York; Phoenix; Washington, DC; and Surrounding Areas.” Journal of the Air & Waste Management Association 58 (12): 1598–1615. doi:10.3155/1047-3289.58.12.1598.
Bobbink, Roland, K. Hicks, J. Galloway, T. Spranger, R. Alkemade, M. Ashmore, M. Bustamante, S. Cinderby, E. Davidson, and F. Dentener. 2010. “Global Assessment of Nitrogen Deposition Effects on Terrestrial Plant Diversity: A Synthesis.” Ecological Applications 20 (1): 30–59.
Brooks, Matthew L. 2003. “Effects of Increased Soil Nitrogen on the Dominance of Alien Annual Plants in the Mojave Desert.” Journal of Applied Ecology 40 (2): 344–53. doi:10.1046/j.1365-2664.2003.00789.x.
178
Bytnerowicz, Andrzej, Maria Jose Sanz, Michael J. Arbaugh, Pamela E. Padgett, David P. Jones, and Antonio Davila. 2005. “Passive Sampler for Monitoring Ambient Nitric Acid (HNO3) and Nitrous Acid (HNO2) Concentrations.” Atmospheric Environment 39 (14): 2655–60. doi:10.1016/j.atmosenv.2005.01.018.
Cisneros, Ricardo, Andrzej Bytnerowicz, Donald Schweizer, Sharon Zhong, Samuel Traina, and Deborah H. Bennett. 2010. “Ozone, Nitric Acid, and Ammonia Air Pollution Is Unhealthy for People and Ecosystems in Southern Sierra Nevada, California.” Environmental Pollution 158 (10): 3261–71. doi:10.1016/j.envpol.2010.07.025.
Clark, Christopher M., and David Tilman. 2008. “Loss of Plant Species after Chronic Low-Level Nitrogen Deposition to Prairie Grasslands.” Nature 451 (7179): 712– 15. doi:10.1038/nature06503.
Coutts, Andrew M., Jason Beringer, and Nigel J. Tapper. 2007. “Characteristics Influencing the Variability of Urban CO2 Fluxes in Melbourne, Australia.” Atmospheric Environment 41 (1): 51–62. doi:10.1016/j.atmosenv.2006.08.030.
Day, Thomas A, Patricia Gober, Fusheng S Xiong, and Elizabeth A Wentz. 2002. “Temporal Patterns in near-Surface CO2 Concentrations over Contrasting Vegetation Types in the Phoenix Metropolitan Area.” Agricultural and Forest Meteorology 110 (3): 229–45. doi:10.1016/S0168-1923(01)00288-X.
Delon, C., C. Galy-Lacaux, M. Adon, C. Liousse, D. Serça, B. Diop, and A. Akpo. 2012. “Nitrogen Compounds Emission and Deposition in West African Ecosystems: Comparison between Wet and Dry Savanna.” Biogeosciences 9 (1): 385–402. doi:10.5194/bg-9-385-2012.
Dentener, F., J. Drevet, J. F. Lamarque, I. Bey, B. Eickhout, A. M. Fiore, D. Hauglustaine, et al. 2006. “Nitrogen and Sulfur Deposition on Regional and Global Scales: A Multimodel Evaluation.” Global Biogeochemical Cycles 20 (4): GB4003. doi:10.1029/2005GB002672.
Duren, Riley M., and Charles E. Miller. 2012. “Measuring the Carbon Emissions of Megacities.” Nature Climate Change 2 (8): 560–62. doi:10.1038/nclimate1629.
Fenn, Mark E., E. B. Allen, S. B. Weiss, S. Jovan, L. H. Geiser, G. S. Tonnesen, R. F. Johnson, et al. 2010. “Nitrogen Critical Loads and Management Alternatives for N-Impacted Ecosystems in California.” Journal of Environmental Management 91 (12): 2404–23. doi:10.1016/j.jenvman.2010.07.034.
Galloway, James N., Alan R. Townsend, Jan Willem Erisman, Mateete Bekunda, Zucong Cai, John R. Freney, Luiz A. Martinelli, Sybil P. Seitzinger, and Mark A. Sutton.
179
2008. “Transformation of the Nitrogen Cycle: Recent Trends, Questions, and Potential Solutions.” Science 320 (5878): 889–92.
García, Ma Ángeles, Ma Luisa Sánchez, and Isidro A. Pérez. 2012. “Differences between Carbon Dioxide Levels over Suburban and Rural Sites in Northern Spain.” Environmental Science and Pollution Research 19 (2): 432–39. doi:10.1007/s11356-011-0575-4.
George, K., L. H. Ziska, J. A. Bunce, and B. Quebedeaux. 2007. “Elevated Atmospheric CO2 Concentration and Temperature across an Urban–rural Transect.” Atmospheric Environment 41 (35): 7654–65. doi:10.1016/j.atmosenv.2007.08.018.
Gregg, Jillian W., Clive G. Jones, and Todd E. Dawson. 2003. “Urbanization Effects on Tree Growth in the Vicinity of New York City.” Nature 424 (6945): 183–87. doi:10.1038/nature01728.
———. 2006. “PHYSIOLOGICAL AND DEVELOPMENTAL EFFECTS OF O3 ON COTTONWOOD GROWTH IN URBAN AND RURAL SITES.” Ecological Applications 16 (6): 2368–81. doi:10.1890/1051- 0761(2006)016[2368:PADEOO]2.0.CO;2.
Grimmond, C. S. B, T. S King, F. D Cropley, D. J Nowak, and C Souch. 2002. “Local- Scale Fluxes of Carbon Dioxide in Urban Environments: Methodological Challenges and Results from Chicago.” Environmental Pollution 116, Supplement 1 (March): S243–S254. doi:10.1016/S0269-7491(01)00256-1.
Gurney, Kevin R., Daniel L. Mendoza, Yuyu Zhou, Marc L. Fischer, Chris C. Miller, Sarath Geethakumar, and Stephane de la Rue du Can. 2009. “High Resolution Fossil Fuel Combustion CO2 Emission Fluxes for the United States.” Environmental Science & Technology 43 (14): 5535–41. doi:10.1021/es900806c.
Hall, Sharon J., Ryan A. Sponseller, Nancy B. Grimm, David Huber, Jason P. Kaye, Christopher Clark, and Scott L. Collins. 2011. “Ecosystem Response to Nutrient Enrichment across an Urban Airshed in the Sonoran Desert.” Ecological Applications 21 (3): 640–60. doi:10.1890/10-0758.1.
Helfter, C., D. Famulari, G. J. Phillips, J. F. Barlow, C. R. Wood, C. S. B. Grimmond, and E. Nemitz. 2011. “Controls of Carbon Dioxide Concentrations and Fluxes above Central London.” Atmos. Chem. Phys. 11 (5): 1913–28. doi:10.5194/acp- 11-1913-2011.
Hidy, George M., and William T. Pennell. 2010. “Multipollutant Air Quality Management.” Journal of the Air & Waste Management Association 60 (6): 645–
180
74. doi:10.3155/1047-3289.60.6.645.
Holland, Elisabeth A., Bobby H. Braswell, James Sulzman, and Jean-Francois Lamarque. 2005. “NITROGEN DEPOSITION ONTO THE UNITED STATES AND WESTERN EUROPE: SYNTHESIS OF OBSERVATIONS AND MODELS.” Ecological Applications 15 (1): 38–57. doi:10.1890/03-5162.
Idso, Craig D, Sherwood B Idso, and Robert C Balling Jr. 2001. “An Intensive Two- Week Study of an Urban CO2 Dome in Phoenix, Arizona, USA.” Atmospheric Environment 35 (6): 995–1000. doi:10.1016/S1352-2310(00)00412-X.
Idso, Craig D., Sherwood B. Idso, and Robert C. Balling Jr. 1998. “The Urban Co2 Dome of Phoenix, Arizona.” Physical Geography 19 (2): 95–108. doi:10.1080/02723646.1998.10642642.
Karnosky, David F., John M. Skelly, Kevin E. Percy, and Art H. Chappelka. 2007. “Perspectives Regarding 50 Years of Research on Effects of Tropospheric Ozone Air Pollution on US Forests.” Environmental Pollution 147 (3): 489–506. doi:10.1016/j.envpol.2006.08.043.
Koerner, B., and J. Klopatek. 2002. “Anthropogenic and Natural CO2 Emission Sources in an Arid Urban Environment.” Environmental Pollution 116, Supplement 1 (March): S45–S51. doi:10.1016/S0269-7491(01)00246-9.
Koutrakis, P., C. Sioutas, S. T. Ferguson, J. M. Wolfson, James D. Mulik, and Robert M. Burton. 1993. “Development and Evaluation of a Glass Honeycomb Denuder/filter Pack System to Collect Atmospheric Gases and Particles.” Environmental Science & Technology 27 (12): 2497–2501. doi:10.1021/es00048a029.
Lee, Sang-Mi, H. J. S. Fernando, and S. Grossman-Clarke. 2007. “MM5-SMOKE- CMAQ as a Modeling Tool for 8-H Ozone Regulatory Enforcement: Application to the State of Arizona.” Environmental Modeling & Assessment 12 (1): 63–74. doi:10.1007/s10666-006-9053-7.
Li, Kaihui, Xuejun Liu, Wei Song, Yunhua Chang, Yukun Hu, and Changyan Tian. 2013. “Atmospheric Nitrogen Deposition at Two Sites in an Arid Environment of Central Asia.” PLoS ONE 8 (6): e67018. doi:10.1371/journal.pone.0067018.
Lohse, Kathleen A., Diane Hope, Ryan Sponseller, Jonathan O. Allen, and Nancy B. Grimm. 2008. “Atmospheric Deposition of Carbon and Nutrients across an Arid Metropolitan Area.” Science of The Total Environment 402 (1): 95–105. doi:10.1016/j.scitotenv.2008.04.044.
181
Long, Stephen P, Elizabeth A Ainsworth, Alistair Rogers, and Donald R Ort. 2004. “Rising Atmospheric Carbon Dioxide: Plants FACE the Future.” Annual Review of Plant Biology 55: 591–628. doi:10.1146/annurev.arplant.55.031903.141610.
Lovett, Gary M. 1994. “Atmospheric Deposition of Nutrients and Pollutants in North America: An Ecological Perspective.” Ecological Applications 4 (4): 629–50.
Lovett, Gary M., Timothy H. Tear, David C. Evers, Stuart E.G. Findlay, B. Jack Cosby, Judy K. Dunscomb, Charles T. Driscoll, and Kathleen C. Weathers. 2009. “Effects of Air Pollution on Ecosystems and Biological Diversity in the Eastern United States.” Annals of the New York Academy of Sciences 1162 (1): 99–135. doi:10.1111/j.1749-6632.2009.04153.x.
Melillo, Jm, Ad Mcguire, Dw Kicklighter, B. Moore, Cj Vorosmarty, and Al Schloss. 1993. “Global Climate-Change and Terrestrial Net Primary Production.” Nature 363 (6426): 234–40. doi:10.1038/363234a0.
Molina, Mario J., and Luisa T. Molina. 2004. “Megacities and Atmospheric Pollution.” Journal of the Air & Waste Management Association 54 (6): 644–80. doi:10.1080/10473289.2004.10470936.
Monks, P. S., C. Granier, S. Fuzzi, A. Stohl, M. L. Williams, H. Akimoto, M. Amann, et al. 2009. “Atmospheric Composition Change – Global and Regional Air Quality.” Atmospheric Environment, ACCENT Synthesis, 43 (33): 5268–5350. doi:10.1016/j.atmosenv.2009.08.021.
Nasrallah, Hassan A, Robert C Balling Jr, Shaker Mohammed Madi, and Lamya Al- Ansari. 2003. “Temporal Variations in Atmospheric CO2 Concentrations in Kuwait City, Kuwait with Comparisons to Phoenix, Arizona, USA.” Environmental Pollution 121 (2): 301–5. doi:10.1016/S0269-7491(02)00221-X.
Nunnermacker, L. J., J. Weinstein-Lloyd, L. Kleinman, P. H. Daum, Y. N. Lee, S. R. Springston, P. Klotz, L. Newman, G. Neuroth, and P. Hyde. 2004. “Ground- Based and Aircraft Measurements of Trace Gases in Phoenix, Arizona (1998).” Atmospheric Environment 38 (29): 4941–56. doi:10.1016/j.atmosenv.2004.04.033.
Pan, Y. P., Y. S. Wang, G. Q. Tang, and D. Wu. 2012. “Wet and Dry Deposition of Atmospheric Nitrogen at Ten Sites in Northern China.” Atmos. Chem. Phys. 12 (14): 6515–35. doi:10.5194/acp-12-6515-2012.
Paoletti, Elena, and William J. Manning. 2007. “Toward a Biologically Significant and Usable Standard for Ozone That Will Also Protect Plants.” Environmental Pollution 150 (1): 85–95. doi:10.1016/j.envpol.2007.06.037.
182
Pardo, Linda H., Mark E. Fenn, Christine L. Goodale, Linda H. Geiser, Charles T. Driscoll, Edith B. Allen, Jill S. Baron, et al. 2011. “Effects of Nitrogen Deposition and Empirical Nitrogen Critical Loads for Ecoregions of the United States.” Ecological Applications 21 (8): 3049–82. doi:10.1890/10-2341.1.
Parrish, David D., Hanwant B. Singh, Luisa Molina, and Sasha Madronich. 2011. “Air Quality Progress in North American Megacities: A Review.” Atmospheric Environment 45 (39): 7015–25. doi:10.1016/j.atmosenv.2011.09.039.
Pataki, D. E., R. J. Alig, A. S. Fung, N. E. Golubiewski, C. A. Kennedy, E. G. Mcpherson, D. J. Nowak, R. V. Pouyat, and P. Romero Lankao. 2006. “Urban Ecosystems and the North American Carbon Cycle.” Global Change Biology 12 (11): 2092–2102. doi:10.1111/j.1365-2486.2006.01242.x.
Pataki, D. E., T. Xu, Y. Q. Luo, and J. R. Ehleringer. 2007. “Inferring Biogenic and Anthropogenic Carbon Dioxide Sources across an Urban to Rural Gradient.” Oecologia 152 (2): 307–22. doi:10.1007/s00442-006-0656-0.
Payne, Richard J., Nancy B. Dise, Carly J. Stevens, David J. Gowing, Cecilia Duprè, Edu Dorland, Cassandre Gaudnik, et al. 2013. “Impact of Nitrogen Deposition at the Species Level.” Proceedings of the National Academy of Sciences 110 (3): 984– 87. doi:10.1073/pnas.1214299109.
Payne, Richard J., Carly J. Stevens, Nancy B. Dise, David J. Gowing, Michael G. Pilkington, Gareth K. Phoenix, Bridget A. Emmett, and Michael R. Ashmore. 2011. “Impacts of Atmospheric Pollution on the Plant Communities of British Acid Grasslands.” Environmental Pollution, Nitrogen Deposition, Critical Loads and Biodiversity, 159 (10): 2602–8. doi:10.1016/j.envpol.2011.06.009.
Pope, Ronald, and Jianguo Wu. 2014. “Characterizing Air Pollution Patterns on Multiple Time Scales in Urban Areas: A Landscape Ecological Approach.” Urban Ecosystems, 1–20. Accessed March 31. doi:10.1007/s11252-014-0357-0.
Rao, Leela E., Edith B. Allen, and Thomas Meixner. 2010. “Risk-Based Determination of Critical Nitrogen Deposition Loads for Fire Spread in Southern California Deserts.” Ecological Applications 20 (5): 1320–35. doi:10.1890/09-0398.1.
Roadman, M. J, J. R Scudlark, J. J Meisinger, and W. J Ullman. 2003. “Validation of Ogawa Passive Samplers for the Determination of Gaseous Ammonia Concentrations in Agricultural Settings.” Atmospheric Environment 37 (17): 2317–25. doi:10.1016/S1352-2310(03)00163-8.
Seto, Karen C., Burak Güneralp, and Lucy R. Hutyra. 2012. “Global Forecasts of Urban Expansion to 2030 and Direct Impacts on Biodiversity and Carbon Pools.”
183
Proceedings of the National Academy of Sciences 109 (40): 16083–88. doi:10.1073/pnas.1211658109.
Smith, Stanley D., Therese N. Charlet, Stephen F. Zitzer, Scott R. Abella, Cheryl H. Vanier, and Travis E. Huxman. 2014. “Long-Term Response of a Mojave Desert Winter Annual Plant Community to a Whole-Ecosystem Atmospheric CO2 Manipulation (FACE).” Global Change Biology 20 (3): 879–92. doi:10.1111/gcb.12411.
Song, Tao, and Yuesi Wang. 2012. “Carbon Dioxide Fluxes from an Urban Area in Beijing.” Atmospheric Research 106 (March): 139–49. doi:10.1016/j.atmosres.2011.12.001.
Sponseller, Ryan A., Sharon J. Hall, David P. Huber, Nancy B. Grimm, Jason P. Kaye, Christopher M. Clark, and Scott L. Collins. 2012. “Variation in Monsoon Precipitation Drives Spatial and Temporal Patterns of Larrea Tridentata Growth in the Sonoran Desert.” Functional Ecology 26 (3): 750–58. doi:10.1111/j.1365- 2435.2012.01979.x.
Templer, Pamela H. 2013. “Biogeochemistry: Limits on Carbon Uptake by Plants.” Nature Climate Change 3 (3): 184–85. doi:10.1038/nclimate1841.
United Nations. 2012. “World Urbanization Prospects, The 2011 Revision”. United Nations publication.
Vourlitis, George L. 2012. “Aboveground Net Primary Production Response of Semi- Arid Shrublands to Chronic Experimental Dry-Season N Input.” Ecosphere 3 (3): art22. doi:10.1890/ES11-00339.1.
Wang, Ge, and Martin Ostoja-Starzewski. 2004. “Influence of Topography on the Phoenix CO2 Dome: A Computational Study.” Atmospheric Science Letters 5 (5): 103–7. doi:10.1002/asl.67.
Watson, John G., Judith C. Chow, Fred W. Lurmann, and Stefan P. Musarra. 1994. “Ammonium Nitrate, Nitric Acid, and Ammonia Equilibrium in Wintertime Phoenix, Arizona.” Air & Waste 44 (4): 405–12. doi:10.1080/1073161X.1994.10467262.
Wentz, Elizabeth A., Patricia Gober, Robert C. Balling, and Thomas A. Day. 2002. “Spatial Patterns and Determinants of Winter Atmospheric Carbon Dioxide Concentrations in an Urban Environment.” Annals of the Association of American Geographers 92 (1): 15–28. doi:10.1111/1467-8306.00277.
Wesely, M. L, and B. B Hicks. 2000. “A Review of the Current Status of Knowledge on
184
Dry Deposition.” Atmospheric Environment 34 (12–14): 2261–82. doi:10.1016/S1352-2310(99)00467-7.
Zhang, L., J. R. Brook, and R. Vet. 2003. “A Revised Parameterization for Gaseous Dry Deposition in Air-Quality Models.” Atmospheric Chemistry and Physics 3 (6): 2067–82.
185 the metropolitan Phoenix, region AZ (urban) and of in outlying native ecosystems to the east the and city. west of 1 TABLE
§)39th)Ave)is)the)calibration)site)located)in)dense)residential)neighborhood)and)only)site)not)located)within)protected)native)ecosystem ¶)Mean)annual)precipitation)average)for)2006)^)2013)(FCDMC)met)station)data)Appendix)2) #Temperature,)relative)humidty)and)precipitation)averaged)for)summer)sampling)periods)(FCDMC)met)station)data)Appendix)2)) ))))based)on)10)km)buffer)surrounding)each)site)location.)) ^Traffic)(average)weekday)traffic/mile)of)road)from)MAG)Transport)Division))and)housing)density)(average)households/km2)from)2010)Census)block))are) in to city calculated was *Distance of center Phoenix, to urban from site location distance Euclidean ArcGISon 9.3 based AZ Ozone)and)nitrogen)concentrations)were)monitored)at)all)locations.)Carbon)dioxide)was)monitored)at)a)subset)of)sites)(EME,)DBG,)and)MCS) Outlying)East Urban Outlying)West Location( Table(1: )Site)characteristics)within)open)space)preserves)in)metropolitan)region)of)Phoenix,)AZ)(urban))and)in)outlying)native)ecosystem)to)the)east)and)west)of)the)city.) :
Site characteristics within open space pres Lost)Dutchman)Park)(LDP) McDowell)Sonoran)Preserve)(MSP) McDowell)Mountain)South)(MCS) Salt)River)Recreation)area)(SRR) 39th)Ave)&)Earll)Dr)(39th)Ave)§ South)Mountain)West)(SMW) Piestawa)Peak)(PWP) Desert)Botanical)Garden)(DBG) White)Tank)Mountain)(WTM) Estrella)Mountain)East)(EME) Site(Name
Elevation( (m) 620 514 539 434 336 458 456 396 454 331 erves in Phoenix, Arizona. Site characteristics within open space preserves in Distance(to( city(center( (km)* 65 30 45 56 12 15 14 51 29 7 density^( Traffic( 21833 45169 12310 64899 27614 54661 76219 12428 1940 7460 density^(((( Housing( 182 725 212 658 615 29 35 63 49 2 temperature( Mean(2013(( summer( 31.3 31.1 32.7 31.6 33.8 35.7 33.1 33.5 33.6 33.8 (C)# Mean(2013( humidity( summer( relative( 27.0 24.7 23.3 28.9 24.9 20.2 22.8 24.0 21.2 20.0 (%)# precipitation( Total(2013( summer( (mm)# 99.6 37.6 15.1 10.2 27.9 10.2 37.6 17.8 24.6 10.2
Mean(annual( precipitation( (mm)¶ 211.4 215.6 192.0 201.5 161.7 168.3 152.4 154.8 145.3 133.2
186 meansdifferent among sites (pairwise comparison post NH 2 TABLE
3 , NO sampling)periods.))Different)letters)represent)significantly)different)means)among)sites)(pairwise)comparison)post3hoc)test)with)Bonferroni)adjustment).) Differences)in)N)among)sites)were)tested)with)non3parametric)Kruskal)Wallis,)and)ozone)was)tested)with)parametric)ANOVA. Total)N)is)a)sum)of)HNO3,)NH3,)and)NOx.))If)a)compound)was)not)measured)during)particular)period,)the)corresponding)site)average)was)used)to)calculate)Total)N Site)names)as)in)Table)1;)SMW)values)are)from)the)middle)site)along)transect)(SMW)T3) Outlying)East Urban Outlying)West Location# Table#2: x , and total N : Mean site O )Range)and)two3three)week)mean)(SD))ozone,)nitric)acid,)ammonia,)nitrogen)oxides,and))total)N)(HNO3)+)NH3)+)NOx))concentrations)during)summer)2013) LDP MSP MCS SRR 39th)Ave SMW PWP DBG EME WTM Site#Name ( HNO 3 p3value , HNO 3
O3#(ppb) 42.3))3)56.0 45.0)3)65.4 41.6)3)51.7 32.4)3)41.4 41.2)3)53.8 40.4)3)51.8 31.9)3)46.8 40.6)3)50.7 39.0)3)50.0 + 3 44)3)55.8
Range , NH NH 3 3
+ , NO
NO 47.7)(5.5)ab 49.2)(6.0)ab 45.9)(3.7)ab 46.2)(5.1)ab 46.5)(4.2)ab 40.5)(6.3)ab 46.1)(4.0)ab 44.9)(4.8)ab Mean#(SD) 50.0)(8.7)b 37.9)(4.6)a x x , and total N ) 0.04 concentrationsg m ( µ HNO3#(ug#N#/#m3) 0.4)3)1.0 0.4)3)0.6 0.4)3)0.8 0.4)3)0.9 0.4)3)1.1 0.4)3)1.2 0.5)3)1.3 0.5)3)1.5 0.3)3)0.8 0.3)3)0.8 Range - h concentrationsg m ( µ oc test). Mean#(SD) 0.7)(0.3)a 0.5)(0.1)a 0.6)(0.2)a 0.6)(0.2)a 0.7)(0.3)a 0.8)(0.3)a 0.9)(0.3)a 1.0)(0.4)a 0.5)(0.2)a 0.6)(0.2)a 0.3 - 3
)
NH3#(ug#N#/#m3) during summerletters 2013. Different represent significantly 6.9)3)14.1 4.9)3)13.1 6.3)3)10.9 2.6)3)6.0 5.8)3)6.8 3.2)3)5.0 2.4)3)4.7 4.8)3)8.1 4.9)3)8.6 2.6)3)9.5 Range - 3 Mean#(SD) 10.9)(3.7)a ) 4.5)(1.7)a 6.3)(0.7)a 4.1)(0.9)a 3.8)(1.2)a 9.0)(4.1)a 6.6)(1.7)a 8.7)(2.3)a 6.9)(1.9)a 5.7)(3.5)a . Range and two 0.06 NOx#(ug#N#/#m3) 0.1)3)0.4 0.5)3)0.5 0.1)3)0.3 0.2)3)0.4 1.5)3)1.9 0.2)3)0.7 0.5)3)0.6 1.1)3)1.2 0.2)3)0.3 0.1)3)0.3 Range - three weeks mean (SD) Mean#(SD) 0.5)(NA)ab 0.4)(0.2)ab 0.5)(0.1)ab 1.1)(0.1)bc 0.3)(0.2)a 0.2)(0.1)a 0.3)(0.2)a 1.7)(0.2)c 0.2)(0.1)a 0.2)(0.1)a 0.09 Total#N#(ug#N#/#m3) 9.0)3)16.6 5.9)3)14.8 8.2)3)13.4 3.1)3)10.5 3.3)3)7.0 6.9)3)7.9 4.0)3)6.1 3.1)3)5.6 5.9)3)9.4 5.5)3)9.7 Range 10.3)(3.2)bc 10.8)(1.9)bc Mean#(SD) 7.3)(0.5)ab 13.2)(2.8)c 8.0)(1.3)ab 7.7)(1.5)ab 6.5)(2.6)ab 5.4)(1.4)a 5.0)(0.7)a 4.6)(0.9)a <0.001 O 3 , HNO 3 187 , method using measured concentrations N and average Sonoran Desert deposition velocities. NO 3 TABLE
x , and total N summer)2013)sampling)periods.))Deposition)was)estimated)by)the)inferential)method.)) Total)N)calculated)as)described)in)Table)2 Site)names)as)in)Table)1;)SMW)values)are)from)the)middle)site)along)transect)(SMW)T3) Outlying)East Urban Outlying)West Location% Table%3:
Mean site HNO )Range)and)two1three)week)mean)(SD))nitric)acid,)ammonia,)nitrogen)oxides,)total)N)(HNO3)+)NH3)+)NOx))deposition)during) ( HNO LDP MSP MCS SRR 39th)Ave SMW PWP DBG EME WTM Site%Name 3
+
3 NH , NH
3
+ 3 HNO3%(kg%N%ha%yr) , NO
NO 1.5)1)3.8 1.6)1)2.4 1.4)1)3.1 1.3)1)3.3 1.4)1)4.3 1.6)1)4.4 1.8)1)4.8 1.9)1)5.5 1.2)1)3.1 1.1)1)3.0 Range x x ) deposition , and total N Mean%(SD) 2.5)(1.0) 2.0)(0.4) 2.3)(0.7) 2.2)(0.8) 2.7)(1.1) 3.0)(1.1) 3.2)(1.3) 3.9)(1.5) 2.0)(0.8) 2.2)(0.4)
( k deposition g N ha N NH3%(kg%N%ha%yr) 1.2)1)2.2) 1.2)1)2.8 2.8)1)3.2 1.5)1)2.4 3.3)1)6.7 2.3)1)6.2 2.3)1)3.8 3.0)1)5.2 2.3)1)4.1 1.2)1)4.5 Range - 1
yr
( - k 1 g )
during summer 2013. Deposition estimated was with the inferential N ha N Mean%(SD) 2.1)(0.8) 3.0)(0.3) 1.9)(0.4) 1.8)(0.6) 5.1)(1.7) 4.3)(1.9) 3.1)(0.8) 4.1)(1.1) 3.3)(0.9) 2.7)(1.7) - 1
yr - 1 ) . Range and two NOx%(kg%N%ha%yr) 0.04)1)0.1 0.1)1)0.1 0.1)1)0.1 0.1)1)0.1 0.1)1)0.1 0.6)1)0.7 0.1)1)0.3 0.2)1)0.2 0.4)1)0.5 0.1)1)0.1 Range Mean%(SD) 0.1)(0.03) 0.1)(0.02) 0.1)(0.04) 0.7)(0.09) 0.2)(0.01) 0.2)(0.01) 0.5)(0.02) 0.1)(0.03) 0.1)(0.03) 0.1)(NA)
- three weeks mean (SD) Total%N%(kg%N%ha%yr) 5.4)1)10.4 4.8)1)10.3 6.3)1)11.1 2.9)1)6.2 4.8)1)5.7 3.4)1)5.5 2.7)1)5.1 4.9)1)8.2 3.6)1)7.2 2.4)1)7.6 Range Mean%(SD) 4.7)(1.4) 5.2)(0.5) 4.3)(0.8) 4.1)(1.0) 8.5)(2.1) 7.5)(2.2) 6.5)(1.5) 8.5)(2.1) 5.3)(1.4) 4.9)(1.9) HNO 3 , NH 3 , 188 interval). reported by location along the transect (n = 3 sampling intervals / location) and by sampling interval (n = 5 locations / sam meters along a 1500 meter transect the from exterior (Point 1) to int Range O and mean of (SD) 4 TABLE
sampling)periods)/)location))and)by)sampling)interval)(n)=)5)locations)/)sampling)interval).) 1500)m)transect)from)the)exterior)(Point)1)))to)interior)(Point)5))of)large)protected)area)within)the)city.))Summer)2013)are)reported)by)location)along)the)transect)(n)=)3) *)Ozone)and)N)sampling)only)occurred)along)the)transect)during)the)final)three)sampling)periods)of)Summer)2013. Sampling)Dates* Transect)locations Table#4: )Range)and)mean)(SD))of)ozone)concentrations,)and)nitric)acid,)ammonia,)nitrogen)oxides,)and)total)N)(HNO3)+)NH3)+)NOx))deposition)sampled)every)300)m)along)a)
: O 3 ,
HNO 30)July)5)14)August 12)July)5)30)July 11)June)5)12)July Interior,)Site)5 Site)4 Site)3 Site)2 Exterior,)Site)1 3 , NH 3 3 , NO , HNO O3#(ppb) 41.23)5)53.8 x 40.0)5)41.4 43.9)5)46.1 49.5)5)53.8 40.9)5)49.5 41.4)5)52.2 40.3)5)50.1 40.0)5)52.1 , and total N Range 3 , NH 3 , NO Mean#(SD) 40.7)(0.6) 51.5)(1.7) 45.4)(4.3) 46.5)(5.4) 47.0)(6.3) 45.3)(4.8) 45.3)(6.2) 45.3(0.9) x deposition , and total N HNO3#(kg#N#ha#yr) 1.5)5)1.7 2.1)5)3.8 1.5)5)3.9 1.5)5)2.1 1.5)5)3.0 1.6)5)3.8 1.5)5)2.3 1.7)5)3.9 Range
( k g ( Mean#(SD) HNO N ha N 1.6)(0.1) 2.5)(0.7) 2.6)(1.0) 1.7)(0.3) 2.2)(0.7) 2.8)(1.1) 1.8)(0.5) 2.6)(1.1) - 3 1
+ erior large (Point 5) of protected area within the city. Data are yr
NH NH3#(kg#N#ha#yr) - 1 5.3)5)6.5 4.4)5)6.7 5.5)5)6.7 4.5)5)5.2 4.4)5)6.2 4.4)5)6.5 4.6)5)5.7 ) Range
along edge ND 3
+
NO Mean#(SD) x 5.8)(0.5) 4.9)(1.0) 6.1)(0.8) 4.9)(0.5) 5.3)(1.3) 5.4)(1.5) 5.2)(0.8) ) deposition ND - to - interior transect protected of urban open space. NOx#(kg#N#ha#yr) 0.1)5)0.2 0.2)5)0.3 0.1)5)0.1 0.1)5)0.2 0.2)5)0.2 0.1)5)0.3 0.2)5)0.2 0.1)5)0.3 Range
( k g N ha N Mean#(SD) 0.2)(0.03) 0.3)(0.06) 0.1)(0.01) 0.2)(0.06) 0.2)(0.05) 0.2)(NA) 0.2)(0.1) 0.2)(NA) - 1
yr - 1 ) Total#N#(kg#N#ha#yr)
7.6)5)10.3 6.0)5)10.3 6.0)5)7.8 6.1)5)8.6 7.8)5)8.3 6.6)5)7.7 6.1)5)9.0 7.1)5)8.6 measured every 300 Range Mean#(SD) 6.9)(0.7) 8.6)(1.1) 7.6)(1.0) 8.0)(0.3) 7.3)(0.6) 7.9)(2.2) 7.4)(1.4) 7.9)(0.8) pling
189
Outlying West Urban Outlying East Urban Protected Desert
Figure'1:"Nitrogen"(NH3,"NOx,"and"HNO3),"O3,"and"CO2"monitoring"sites"in"the"metropolitan" region"of"Phoenix,"Arizona.""O "and"N"were"measured"at"all"sites.""Open"symbols"indicate" FIGURE 1: Nitrogen, O3, 3and CO2 monitoring sites in metropolitan region of Phoenix, Arizona.CO2"monitoring"locaEons.""Starred"diamond"indicates"locaEon"of"the"exteriorGtoGinterior" Nitrogen (NH , NO , and HNO ), O , and CO monitoring sites in the urban"open"space"transect.""All"monitoring"sites"are"located"within"protected"Sonoran"3 x 3 3 2 metropolitaDesert"areas,"except"the"urban"calibraEon"site"(hatched"diamond)"which"is"located"in"a"n region of Phoenix, Arizona. O3 and N were measured at all sites. Open symbolsdense"residenEal"neighborhood.""" indicate CO2 monitoring locations. Starred diamond indicates location of the edge-to-interior urban open space transect. All monitoring sites are located with protected Sonoran Desert areas, except the urban calibration site (hatched diamond) which is located in a dense residential neighborhood.
190
A" B" C"
Figure'2:"Field"set*up"for"measuring"gaseous"ambient"NH3,"NOx,"HNO3,"O3"and"CO2" FIGUREconcentra>ons.""(A)"NH 2: Field set3,"-NOupx ,"HNOfor measuring3,"and"O3"passive"samplers"are"installed"2"meters"above" ambient atmospheric gaseous concentrations. ground"under"protec>ve"covers"to"block"direct"sun"and"rain"during"2"week"sampling"intervals."" Field(B)"Infrared"gas"analyzers"(IRGAs)"are"installed"at"2"and"0.5"meters"above"ground"under"larger" set-up for measuring ambient NH3, NOx, HNO3, O3, and CO2 concentrations. (A)
NHprotec>ve"covers"to"con>nuously"monitor"ambient"CO3, NOx, HNO3, and O3 passive samplers were2"concentra>ons.""(C)"View"of"CO installed 2 meters above2"and"N" ground under protectivepassive"sampler"installed"2"meters"above"ground"under"protected"covers"in"the"desert."" covers to block direct sun and rain during 2 week sampling intervals. (B) Infrared gas analyzers (IRGAs) were installed at 2 and 0.5 meters above ground to continuously monitor ambient CO2 concentrations. (C) View of CO2 and N passive samplers installed at 2 meters above ground in a native desert location.
191 Summer 2013 June 11 June 26 July 12 July 30 August 14 410 EME DBG MCS
400 1:"May"17"A" 2:"June"11"A 3:"June"26"A" 4:"July"12"A 5:"July"30"A 400" June"11" June"26" July"12" July"30"" August"14"
2"" 390 " 390" CO2 2m (ppm) 2m CO2 (ppm) 380"380 Mean"CO June 11 June 26 July 12 July 30 August 14
370"37060 60" EME DBG MCS EME DBG MCS EME DBG MCS EME DBG MCS EME DBG MCS Mean"O
June 11 June 26 July 12 July 30 August 14 (ppb) 12.5 12.5"40 40" West ) " 1 West
Urban 3""
yr Urban 1 East 10.0 10.5" East " ) Ozone (ppb) Ozone A1 20 "yr 7.5 20"
A1 7.5"
5.0 5.0" CO2" (kg"N"ha 0 O3" WestUrbanEast WestUrbanEast WestUrbanEast WestUrbanEast WestUrbanEast Mean"dry"N"deposi 192 Summer 2013 400400% % 2 % 390390% a% a% (ppm) EME Mean%CO b% DBG 380380% MCS 60 60% a% CO2 2m (ppm) 2m CO2 ab% b% Mean%O 370 (ppb) 10.010.0%40 EME DBG MCS 40% West % Urban3% % a% ) ) 1 East 31 7.5207.5% 20% Ozone (ppb) Ozone %yr yr 31 1 b% NOxkg %ha 5.05.0%0 b% NH3kg CO2% kgN ( West Urban East HNO3kgO3% 2.52.5% NOx% Mean%dry%N%deposi.on% N (kgN ha (kgN N NH3% HNO3% 0.00 West% Urban% East% West Urban East Figure'4:%Summer%2013%mean%(+/3SE)%dry%N%(HNO3%+%NH3%+%NOx)% 31 31 deposi.on%(kgN%ha %yr ),%O3(triangles,%ppb),%and%CO2%(circles,%ppm%at%2% FIGUREmeters)%in%outlying%west,%urban,%and%outlying%east%preserves.%%Different% 4: Mean summer N deposition, O3, and CO2 in outlying and urban locations. Summer 2013 mean (+/- SE) dry N deposition (HNO + NH + NO , kg N ha-1 yr-1), O leTers%within%a%pollutant%indicate%significantly%different%means.%%3 3 x 3 (triangles, ppb), and CO2 (circles, ppm at 2 meters)% in outlying west, urban, and outlying eastHorizontal%lines%are%for%reference%to%O native open space preserves. Different3%scale. letters within a pollutant indicate significantly different means. Horizontal lines are reference to O3 scale. 193 Summer 2013 EME DBG MCS Summer 2013 2m Summer 2013 2m Summer 2013 2m 475" Outlying"West" 420 Urban" 420 Outlying"East" 420 475 b" b" b" (EME)" b" (DBG)" a" (MCS)" 400 b" b" a" June 400 June 400 June July July a" July 380 August 380 August 380 August 450"450 360 360 360 June July August June July August June July August 425"425 June" June July" July August August" CO2 2m (ppm) 2m CO2 400"400 375"375 'Carbon'dioxide'at'2'meters'(ppm)' 350"350 0"0 1 2 3 4 56" 6 7 8 9101112131415161718192021222312" 18" 0"0 1 2 3 4 56" 6 7 8 9101112131415161718192021222312" 18" 0"0 1 2 3 4 56" 6 7 8 9101112131415161718192021222312" 18" Time'(hour)' FIGUREFigure'5: "Summer"2013"hourly"CO5: Summer hourly2 "concentraGons"(ppm)"at"2"meters"averaged"(+/M"1SE)"by"month"urban and outlying CO2 concentrations at 2 meters above (June"(circle),"July"(square),"and"August"(diamond))"at"EME,"DBG,"and"MCS.""Dashed"lines"are"summer" ground. Summer 2013 hourly CO2 concentrations (ppm) at 2 meters averaged (+/- 1SE) averages"per"site;"black"line"indicates"400"ppm"for"reference."Insets"show"monthly"median,"1st"and" by monthrd (June (circle), July (square), and August (diamond)) at outlying west, urban and 3 "quarGle"(box),"min"and"max"(whiskers)"CO2"by"month"at"each"site.""Different"leZers"indicate" outlyingsignificantly"different"CO east sites. Dashed2"concentraGons"within"a"site."" lines are summer averages per site; black line indicates 400 ppm" for reference. Insets show monthly median, 1st and 3rd quartile (box), minimum and maximum (whiskers) CO2 by month at each site. Different letters indicate significantly different CO2 concentrations within a site. 194 Winter 2013-2014 EME DBG MCS Winter 2m Winter 2m Winter 2m 475 475 475 475"475 Outlying"West" a" Urban" a" Outlying"East" (EME)" 450 (DBG)"Dec 450 a" (MCS)" 450 Dec Dec Jan a" Jan 425 ab" ab" 425 Feb 425 b" Feb ab" Feb Mar ab" b" a" 400 Mar 400 400 Mar 450"450 375 375 375 Dec Jan Feb Mar Dec Feb Mar Dec Jan Feb Mar December" 425"425 December January" January February" February March" March CO2 2m (ppm) 2m CO2 400"400 375"375 %Carbon%dioxide%at%2%meters%(ppm)% 350"350 0"0 1 2 3 4 56" 6 7 8 9101112131415161718192021222312" 18" 0"0 1 2 3 4 56" 6 7 8 9101112131415161718192021222312" 18" 0"0 1 2 3 4 56" 6 7 8 9101112131415161718192021222312" 18" Time%(hour)% Figure%6:"Winter"2013C2014"hourly"CO2"concentraGons"(ppm)"at"2"meters"averaged"(+/C"1SE)"by"month" FIGURE(December"(circle),"January"(square),"February"(diamond),"and"March"(triangle))"at"EME,"DBG,"and"MCS."" 6: Winter hourly urban and outlying CO concentrations at 2 meters above Dashed"lines"are"winter"season"site"average;"black"line"indicates"400"ppm"for"reference.""Insets"show"2 st rd ground.monthly"median,"1 Winter 2013"and"3 -"quarGle"(box),"min"and"max"(whiskers)"CO2014 hourly CO2 concentrations2"by"month"at"each"site.""Different" (ppm) at 2 meters averaged (+/- 1SE)leZers"indicate"significantly"different"CO by month (December (circle),2"concentraGons"within"a"site."" January (square), February (diamond), and March (triangle)) at outlying west, urban and outlying east sites. Dashed lines are winter averages per site; black line indicates 400 ppm for reference. Insets show monthly st rd median, 1 and 3 quartile (box), minimum and maximum (whiskers) CO2 by month at each site. Different letters indicate significantly different CO2 concentrations within a site. 195 Summer 2013 EME DBG MCS Summer 2013 0.5m Summer 2013 0.5m Summer 2013 0.5m 475"475 Outlying"West" 420 Urban" 420 Outlying"East" 420 (EME)" a" a" (DBG)" a" (MCS)" a" 400 400 400 June a" a" June b" June a" July July July 380 August 380 August 380 ab" August 450"450 360 360 360 June July August June July August June July August 425"425 June June" July July" August August" CO2 0.5m (ppm) 0.5m CO2 400"400 375"375 Carbon%Dioxide%at%0.5%meter%(ppm)% 350"350 0"0 1 2 3 4 56" 6 7 8 9101112131415161718192021222312" 18" 0"0 1 2 3 4 56" 6 7 8 9101112131415161718192021222312" 18" 0"0 1 2 3 4 56" 6 7 8 9101112131415161718192021222312" 18" Time%(hour)% Figure%7:"Summer"2013"hourly"CO2"concentraGons"(ppm)"at"0.5"meters"averaged"(+/N"1SE)"by"month" (June"(circle),"July"(square),"and"August"(diamond))"at"EME,"DBG,"and"MCS.""Dashed"lines"are"summer" FIGUREaverages"per"site;"black"line"indicates"400"ppm"for"reference."Insets"show"monthly"median,"1 7: Summer hourly urban and outlying CO2 concentrations at 0.5st"and" meters above rd ground.3 "quarGle"(box),"min"and"max"(whiskers)"CO Summer 2013 hourly CO2 concentrations2"by"month"at"each"site.""Different"leZers"indicate" (ppm) at 0.52 meters averaged (+/- 1SE)significantly"different"CO by month (June (circle),2"concentraGons"within"a"site."" July (square), and August (diamond)) at outlying west, urban and outlying east sites. Dashed lines are summer averages per site; black line indicates 400 ppm for reference. Insets show monthly median, 1st and 3rd quartile (box), minimum and maximum (whiskers) CO2 by month at each site. Different letters indicate significantly different CO2 concentrations within a site. 196 Winter 2013-2014 EME DBG MCS Winter 2m Winter 2m 475 420 Urban" 420 475" Outlying"West" b" ab" Outlying"East" a a a" ab" (DBG)"Dec (MCS)" a a (EME)" 400 400 Dec No#data#Jan Jan Feb 380 380 Feb Mar Mar 450"450 360 360 Dec Jan Feb Mar Dec Jan Feb Mar December" 425"425 January" December January February" February March" March CO2 0.5m (ppm) 0.5m CO2 400"400 375"375 %Carbon%dioxide%at%0.5%meters%(ppm)% 350"350 0"0 1 2 3 4 56" 6 7 8 9101112131415161718192021222312" 18" 0"0 1 2 3 4 56" 6 7 8 9101112131415161718192021222312" 18" 0"0 1 2 3 4 56" 6 7 8 9101112131415161718192021222312" 18" Time%(hour)% Figure%8:"Winter"2013C2014"hourly"CO2"concentraGons"(ppm)"at"0.5"meter"averaged"(+/C"1SE)"by"month" (December"(circle),"January"(square),"February"(diamond),"and"March"(triangle))"at"EME,"DBG,"and"MCS."" FIGUREDashed"lines"are"winter"site"average;"black"line"indicates"400"ppm"for"reference."Insets"show"monthly" 8: Winter hourly urban and outlying CO2 concentrations at 0.5 meters above st rd ground.median,"1 Winter"and"3 "quarGle"(box),"min"and"max"(whiskers)"CO 2013-2014 hourly CO2 concentrations2"by"month"at"each"site.""Different"leZers" (ppm) at 0.5 meters averaged (+/- 1SE)indicate"significantly"different"CO by month (December2 "concentraGons"within"a"site.""There"are"no"data"for"the"urban"site."(circle), January (square), February (diamond), and March (triangle)) at outlying west, urban and outlying east sites. Dashed lines are winter averages per site; black line indicates 400 ppm for reference. Insets show monthly st rd median, 1 and 3 quartile (box), minimum and maximum (whiskers) CO2 by month at each site. Different letters indicate significantly different CO2 concentrations within a site. There were no data recorded at 0.5 meters at the urban site due to sensor error. 197 APPENDIX 1 TABLE 1: Dates of sample deployment and number (n) of sites within each region (outlying west, urban and outlying east) at which each atmospheric compound was monitoredTable.1:)Dates)of)sample)deployment)and)collection)and)number)(n))of)sites)within)each)region)at)which) during summer 2013. each)atmospheric)compound)was)monitored)during)summer)2013.) Deployment. Collection. Carbon. Nitric... Nitrogen. Location date date dioxide Ozone acid Ammonia oxides Outlying)West May)17 June)11 1 2 2 2 ND June)11 June)26 1 2 2 ND 2 June)26 July)12 1 2 2 2 ND July)12 July)30 1 2 2 2 ND July)30 August)14 1 2 2 ND 2 Urban May)17 June)11 1 4 4 4 ND June)11 June)26 1 4 4 ND 4 June)26 July)12 1 8 8 8 3 July)12 July)30 1 8 8 8 3 July)30 August)14 1 8 8 ND 8 Outlying)East May)17 June)11 1 3 3 3 ND June)11 June)26 1 3 3 ND 3 June)26 July)12 1 4 4 4 ND July)12 July)30 1 4 4 4 ND July)30 August)14 1 4 4 ND 4 ND)=)No)data.))Sampling)at)MSP)(outlying)east))and)along)the)SMW)transect)(urban))did)not)occur) during)first)two)summer)periods.))Due)to)limited)sampling)equipment,)ammonia)and)nitrogen)oxides) were)sampled)during)opposite)periods,)except)in)the)urban)location)where)sampling)occurred) 198 TABLE 2: Mean (SD) daily CO2 concentrations (ppm) during each collection period in 2013 and 2014. CO2 concentrations were recorded every 10 minutes at 2 and 0.5 meters above ground at each location (raw data were first averaged by day before calculating Table)2:)Mean%(SD)%daily%carbon%dioxide%concentrations%(ppm)%during%each%collection%period%in%2013%and%2014.%% dailyCarbon%dioxide%concentrations%were%recorded%every%10%minutes%at%2%meters%and%0.5%meter%above%ground%at%each% mean concentration). location%(raw%data%first%averaged%by%day%before%calculating%mean%daily%concentration). Collection)Dates Outlying)West)(EME) Urban)(DBG) Outlying)East)(MCS) 2)m 0.5)m 2)m 0.5)m 2)m 0.5)m June%11%'%June%26%2013 391%(7) ND 393%(7) ND 387%(9) ND June%26%%'%Jul%12%2013 390%(2) 377%(7) 390%(4) 384%(5) 391%(4) 372%(13) July%12%'%July%30%2013 388%(9) 386%(6) ND ND 381%(9) 385%(13) July%30%'%August%14%2013 379%(7) ND 391%(6) 386%(5) 374%(4) 368%(4) August%14%'%September%10%2013 381%(8) 383%(7) 378%(9) 384%(5) 376%(3) 377%(5) September%10%'%October%3%2013 374%(7) 378%(5) 377%(9) 388%(10) 376%(10) 384%(5) October%3%'%November%6%2013 383%(3) 385%(3) 393%(9) ND 380%(5) ND November%6%'%December%9%2013 410%(24) 393%(11) 411%(23) ND 384%(6) 389%(9) December%9%%'%January%7%2014 408%(15) 389%(9) ND ND 385%(7) 391%(7) January%7%'%February%7%2014 ND 383%(5) ND ND 388%(4) 395%(5) February%7%'%March%14%2014 398%(18) 391%(9) 415%(14) ND 386%(9) 392%(6) ND%=%No%data%recoded%during%the%collection%period%due%to%sensor%error. 199 APPENDIX 2 TABLE 1: Flood Control District of Maricopa County (FCDMC) station IDs for the nearest meteorological station to the corresponding N, CO2, or O3 monitoring site. FCDMC meteorological data were downloaded for the study period (summer 2013). If more than one site is listed in a column, data were averaged. carbon(dixoide,(and(ozone(monitoring(site(in(this(study.((FCDMC(meteorological(data(were(downloaded(for(the(study(period(Summer(2013).((If( PWP,(Data(from(Phoenix(Dam(station(only(contain(dates(from(2009(\(2013 DBG(&(LDP,(the(Central(Arizona(Phoenix(Long\term(Ecological(Research(Site((CAP(LTER)(maintains(a(meteorological(tower(at(these(locations CITATION,FOR,DATA East Outlying Urban WestOutlying Location Table,1:, more(than(one(site(is(listed(in(a(column,(data(were(average. Flood(Control(District(of(Maricopa(County((FCDMC)(station(IDs(for(the(nearest(meteorological(station(to(the(corresponding(nitrogen,( (SRR) area Recreation Salt River (MCS) South Mountain McDowell Park (LDP) Lost Dutchman West Mountain South (SMW) (PWP) Peak Piestawa (DBG) Garden Botanical Desert White Tanks Park (WTM) Regional West Mountain Estrella (EMW) East (EME) Mountain Estrella Nitrogen,deposition,monitoring,site, :(Flood(Control(District(of(Maricopa(County(ALERT(System,(http://www.fcd.maricopa.gov/Rainfall/rainfall.aspx Wash,(Cloudburst(Wash Saguaro(Lake,(Granite(Reef Fountain(Hills(Fire(Dept,(Lost(Dog( Apache(Trail,(CAP(LTER S(Mtn(Fan(HQ Phoenix(Dam,(Tatum(Basin(Inflow,( Osborn(@(64th(St,(Salt(River(@( McMicken(Dam,(White(Tanks(Peak,( Estrella(Fan,(Waterman(@( Estrella(Fan,(Gila(River(@116th,( FCDMC,Station,name(s) Priest,(Papago(Park,(CAP(LTER( Rainbow,(Waterman Holly(Acres Dreamy(Draw(Dam White(Tanks(#3,(Ford(Cyn(Wash CAP 4740, 4520, LTER 5430, 5425, 5415 6830, 6880, 6890 6675, CAP6675, LTER Precipitation,,,,, station,ID 4660, 5975 6845, 6860 4565 6525 4635 station,ID humidity, Relative, 6781 5951 6561 4796 4616 5436 6891 6891 Temperature, station,ID 6782 5952 6562 4797 4617 5437 6892 6892 200 CHAPTER 5 SYNERGISTIC NET EFFECTS OF URBAN AIR POLLUTION ON ECOSYSTEMS ABSTRACT Plants and ecosystems are rarely exposed to a single pollutant, yet research on the effects of co-occurring atmospheric compounds is limited. Carbon dioxide (CO2), ozone (O3), and nitrogen (N) deposition are elevated in and around cities, affecting air quality at local to global scales. Despite the ecological relevance of CO2, O3, and N as both stressors and resources for primary producers, their combined impacts at current and future concentrations are unknown. To address this gap, I ask, what is the sensitivity of primary producers to co-occurring CO2, O3, and N? Using the Central Arizona Phoenix Long Term Ecological Research (CAP LTER) site as a model system, I examined the net effect of elevated concentrations of CO2, O3, and N on growth and physiological responses of dominant native (Pectocarya recurvata) and non-native (Schismus arabicus) Sonoran Desert winter herbaceous vegetation. Growth of both species was additive (sum of individual responses) when exposed to two factor combinations (O3 x N, CO2 x N, and CO2 x O3) but synergistic (amplifying, i.e. greater than the sum of individual effects) in response to CO2, O3 and N combined. Elevated CO2 mitigated the negative effects of O3 exposure for the non-native grass but had minimal protective effect against high O3 concentrations for the native forb. Contrary to expectations, elevated N did not have a significant individual influence on the responses of either species. Differences in species’ responses to current and future urban air pollution may lead to long-term 201 changes in plant community composition of protected ecosystems. Overall, these findings highlight the vulnerability of native ecosystems to current and future air pollution over the long term and provide empirical evidence for future policies addressing ecosystem sensitivity to multiple stressors in current and predicted future environmental conditions. 202 INTRODUCTION Atmospheric pollutants generated from human activities can affect air quality and ecosystems at local, regional, and global-scales (Dentener et al. 2006; Grimm et al. 2008; Monks et al. 2009). Though cities are point sources of greenhouse gases and other pollutants, urban-generated atmospheric compounds are not restricted to the political boundaries in which they are regulated or monitored (Akimoto 2003; Monks et al. 2009). Thus, despite their distance from cities, protected native lands are exposed to elevated concentrations of atmospheric carbon dioxide (CO2), reactive nitrogen (N), and ozone (O3) from cities (Cook et al. In prep). Individually, these compounds act as either resources or stressors to primary producers, with potentially cascading effects across trophic levels. Yet little is known about the net impacts of these co-occurring ecologically important compounds on ecosystem structure or function. Unaccounted for synergistic or antagonistic effects of multiple pollutants may explain recent regional and global changes in primary production, carbon and N storage, and community structure (Sala et al. 2000; S. D. Smith et al. 2000; Ollinger et al. 2002; Shaw et al. 2002). With over 65% of the world’s population expected to live in cities by 2050 and urban land cover expected to triple by 2030, ecosystems will increasingly be exposed to elevated CO2, O3, and N deposition with cascading impacts on ecosystem processes (IPCC 2014; Seto, Güneralp, and Hutyra 2012; United Nations 2012). For example, by 2030 reactive N deposition above 10 kg N ha-1 yr-1 is expected to affect up to 35% of Earth’s terrestrial land (Dentener et al. 2006). Elevated reactive N gas emissions and deposition from agriculture and combustion activities alleviate nutrient limitation in 203 terrestrial and aquatic ecosystems, stimulating primary production and altering biogeochemical cycling (Aber et al. 1998; Brooks 2003; Báez et al. 2007; Elser et al. 2007; LeBauer and Treseder 2008). Similarly, CO2—a greenhouse gas and key reactant in photosynthesis—is elevated in and near cities as a result of combustion processes (Idso, Idso, and Balling Jr. 2001; George et al. 2007; Bergeron and Strachan 2011). In some species, elevated CO2 can increase the efficiency of plant N-use and water-use, thus alleviating water stress and increasing production (Long et al. 2004; Housman et al. 2006); though in other species, elevated CO2 has minimal effect on primary producers (Morgan et al. 2004; Dukes et al. 2005). On the other hand, tropospheric O3—a secondary, widely dispersed pollutant—has negative effects on plant physiology, leading to leaf and cellular damage, reduced growth, and early senescence with consequences for community composition (Bytnerowicz et al. 1988; Karnosky et al. 2007; Ainsworth et al. 2012; Lombardozzi, Sparks, and Bonan 2013). Atmospheric concentrations of CO2 and tropospheric O3 are both expected to increase 40% globally by 2100 (Horowitz 2006; IPCC 2014). Plants and ecosystems are rarely exposed to a single pollutant (Fangmeier et al. 2002; Lovett et al. 2009) and the combined effects of pollutants can have cascading affects on plant growth, physiological functioning, community composition, and competition. Current global change research has focused on quantifying interactions between land- cover changes (e.g. biological invasion, grazing, and fire), climate-related variables (e.g., precipitation and temperature), and select atmospheric variables (e.g., CO2 and N deposition; Sala et al. 2000; Zavaleta et al. 2003; Henry et al. 2006). These pioneering studies have shown that ecological responses are likely to be synergistic (amplifying or 204 greater than the sum of individual effects) or antagonistic (canceling or less than sum of individual effects) rather than simply additive (Sala et al. 2000; Shaw et al. 2002; Rustad 2008; but see Zavaleta et al. 2003). However, in order to ensure quantifiable effect sizes in complex systems, experimental treatments are often larger than near-future, or even long-term expected conditions (e.g., additions of 7 g N m-2 yr-1 when atmospheric N -2 -1 deposition is ~0.5 g m yr ; e.g., Zavaleta et al. 2003; Weiss 1999; or + 300 ppm CO2 when expected increases are + 150 ppm CO2; e.g., Shaw et al. 2002). In fact, most ecosystems experience small, incremental changes in atmospheric composition over the long-term (M. D. Smith, Knapp, and Collins 2009), and large pulses do not represent current—or often predicted future—concentrations or gradual rates of change that are occurring worldwide (Luo and Reynolds 1999; Shen et al. 2008). Additionally, large experimental changes in ambient conditions make it difficult to test mechanisms, sensitivity, and feedbacks (e.g., down-regulation, Ashmore 2002). Understanding the plant or ecosystem responses at expected levels is essential to examining threshold responses and informing regional critical loads. Critical loads are the level of pollution at which significant ecological effects are expected to occur from elevated N inputs (Fenn et al. 2010; Pardo et al. 2011). To date, critical loads have been determined only for single pollutants, such as N, and thus do not account for more realistic scenarios of exposure to co-occurring compounds. Examining the sensitivity and critical levels relevant to interacting pollutants in native ecosystems will provide empirical evidence to establish frameworks for protection of ecosystem services in addition to human health (Groffman et al. 2006). 205 Airsheds exposed to elevated CO2 and N deposition are often simultaneously exposed to elevated O3 concentrations. Elevated CO2 can reduce the negative physiological impacts of O3 by minimizing stomatal conductance and intercellular O3 exposure (Cardoso‐Vilhena and Barnes 2001; Reid and Fiscus 2008; Ainsworth et al. 2012). The interactive effects of CO2 and O3 are dependent on plant physiological traits, leading to synergistic or antagonistic eco-physiological responses in different ecosystems (Long et al. 2004; Lindroth 2010). Similarly, the net effects of elevated CO2 and N vary across ecosystems based on plant N uptake, soil moisture, temperature, additional nutrient limitations, and species composition (Shaw et al. 2002; Dukes et al. 2005; Reich et al. 2006; Dijkstra et al. 2010; Langley and Megonigal 2010; Reich and Hobbie 2013). Only a few studies have empirically quantified the ecosystem consequences of elevated CO2, N and O3 together. In well-watered agricultural crops, CO2 mitigates O3 damage, and elevated N has little additional influence (Heagle et al. 1999; Cardoso‐ Vilhena and Barnes 2001). However, agricultural plants have different physiological responses to environmental stressors than wild plants, because they are bred for net production under high water and nutrient conditions (sensu Chapin 1980; Chapin 1991). Modeled ecosystem responses to elevated CO2, N and O3 predict potential future feedbacks in temperate forested ecosystems with increased evapotranspiration and reduced N-induced carbon sinks (Ollinger et al. 2002; Felzer et al. 2009). Yet in a field study across an urban-rural gradient, Gregg and colleagues (2003) found high rural O3 concentrations most strongly influenced plant growth compared to other atmospheric components. Overall, however, net physiological, ecosystem, or community responses to realistic combinations of CO2, N, and O3 are difficult to predict from the currently limited 206 research (Sala et al. 2000; Aber et al. 2002; Krupa 2003; Bassin et al. 2009; Lindroth 2010) and there have been several recent calls for research investigating more realistic scenarios in which ecosystems are exposed to elevated CO2, N, and O3 together (Karnosky et al. 2007; Rustad 2008; Templer 2013). Given the unknown ecological impacts of air pollution in urban and surrounding ecosystems around the world, I asked, what is the sensitivity of primary producers to co- occurring CO2, O3, and N at realistic and expected future concentrations? I modeled the experimental designed based on current and predicted near-future air quality conditions in the metropolitan region of Phoenix, Arizona and the surrounding native desert ecosystem. Dryland ecosystems are predicted to be especially sensitive to regional and global anthropogenic changes (Melillo et al. 1993). For example, elevated CO2 is expected to alleviate water limitation by increasing plant water-use efficiency (Smith et al 1997), which would in turn increase primary production and rates of nutrient cycling. Further, in lower productivity systems where primary producers are often co-limited by more than one resource, the effects of multiple changes may be interactive. Within this context, I examined the net effect of elevated levels of CO2, O3, and N on growth and physiological responses of two dominant herbaceous plant species, a native forb (Pectocarya recurvata) and a non-native graminoid (Schismus arabicus). Ephemeral desert plants are ideal candidates to examine mechanistic dose responses and interacting feedbacks related to multiple pollutants. Winter herbaceous plants respond quickly to small changes in the environment and have a relatively short growing season (Sala and Lauenroth 1982). In addition, winter ephemeral plant production provides multiple ecosystem services, such as attractive wildflowers, food for herbivores, and a ‘vernal 207 dam’ preventing inorganic N losses (Venable and Pake 1999; Hall et al. 2011). Thus, I expected the co-occurring pollutants would have synergistic or antagonistic effects, rather than simply additive responses, on plant growth and physiological parameters. METHODS Experimental design In a six-week, multi-factorial dose-response experiment, I examined the dominant native and non-native Sonoran Desert winter herbaceous plant growth and physiological responses to varying combinations of CO2, O3 and N. Between January and March 2013, I conducted a fumigation experiment at the University of California, Riverside and USFS Fire Lab greenhouse facilities in Riverside, California. Using continuously stirred tank reactor (CSTR) fumigation chambers, I grew plants in full factorial treatment combinations of CO2 (ambient (400 ppm), 550, and 700ppm gaseous fumigation), O3 (ambient (15 ppb), 60, and 100 ppb gaseous fumigation), and N (ambient, +4, +8 kgN ha- 1 yr-1 applied as ammonium nitrate, n = 27 treatments for each species). Individual plants (n = 15-17 individuals per species) were grown without competition as subsamples within each treatment. Two to three individuals of each species were successively harvested every 7-10 days throughout the experiment, and the subsamples were then averaged for each harvest. 208 Study species Winter desert herbaceous plants respond quickly, have a short growing season and are well suited to test short-term feedbacks and sensitivities. Pectocarya recurvata (Boraginaceae), a forb, and Schismus arabicus (Poaceae), a graminoid, are two dominant native and non-native species, respectively, within the Sonoran Desert and Central Arizona-Phoenix Long-term Ecological Research (CAP LTER) site encompassing the Phoenix metropolitan region (CAP LTER species composition database). P. recurvata and S. arabicus differ in key functional traits, such as relative growth rate (RGR) and water use efficiency (WUE) related to stomatal conductance and stress tolerance. For example, P. recurvata is characterized by low RGR and high WUE, whereas S. arabicus tends to have high RGR and low WUE (Huxman et al. 2008; Angert et al. 2009). In addition, these species have been examined previously in studies of their eco- physiological responses and thus methods to account for their small leaf area have been previously established (Huxman et al. 2008; Angert et al. 2009; Gremer et al. 2012). Seeds for each species were collected in Spring 2012 in native Sonoran Desert locations east of Phoenix, Arizona (Salt River Recreation Area or Lost Dutchman Park) and stored in a cool, dark, dry place until the experiment. In December 2012, all seeds were soaked in 200 ppm Gibbarellic Acid (GA3) for 24 hours to stimulate germination. At the end of 24-hour soak, seeds were transferred to small germinating trays and grown in a mix of vermiculite and arid zone potting soil (Sunshine Mix #4, SunGro Horticulture, Agawam, MA). Germination occurred in a greenhouse with natural diurnal light and trays were watered every 2-3 days. Approximately 4 weeks after germination, individual seedlings were transplanted into separate “cone-tainers” (individual plastic cells, Stuewe 209 & Sons, Inc Tangent, OR) to exclude competition from the experiment. An arid zone potting soil (Sunshine Mix #4) was used in all cone-tainers to minimize additional variation. Plants were established in the individual cone-tainers for 5-7 days before beginning experimental treatments. Throughout the six-week fumigation experiment, plants were watered with de-ionized water every 5-7 days. Experimental treatment set-up I used 9 CSTR chambers, each 1.35 m diameter x 1.35 m height (Figure 1) located within one greenhouse for simultaneous exposure of plants under similar temperature, humidity, and air-flow conditions (Padgett et al. 2004). With a continuously moving impeller and exchange of air, the CSTR chambers ensure even mixing of gases throughout the chamber. I used a gas delivery and monitoring system to regulate diurnal fluctuations of O3 and maintain a consistent daily CO2 concentration in each chamber. Carbon dioxide and ozone fumigation: Carbon dioxide was distributed directly to each chamber from a 50 lb pressurized CO2 gas canister regulated by a flow meter and manifold system. Chambers receiving elevated CO2 concentrations each contained an infrared gas analyzer (IRGA) with an integrated relay system (iSense CM-0043, CO2meter, Ormand Beach, Fl) and solenoid valve to regulate CO2 gas flow into the chamber. Each chamber’s IRGA was programmed to maintain the specified CO2 concentration by measuring CO2 within the chamber every 10 seconds and opening or closing the flow valve as needed to meet the specified range of CO2. Three chambers were fumigated with approximately 700 ppm CO2 (daily average for high CO2 treatments), three chambers with 550 ppm CO2 (daily average for mid-level CO2 210 treatments), and three chambers received no CO2 inputs for ambient concentrations approximately 400 ppm CO2 (low-level ambient CO2 treatments, Table 1). Ozone was generated from charcoal filtered air by electrical discharge (Griffin O3 synthesizer, Griffin Technics Corp Lodi, NJ). Oxygen was released between 9:00 and 18:00 daily into electrostatic ozone generator and then distributed through a manifold system to each chamber at a specified O3 concentration. Each chamber concentration was adjusted via flow meters such that desired peak O3 concentration remained stable throughout the day. Three chambers were set up to be fumigated with approximately 100 ppb O3 (daytime average for high O3 treatment), three chambers with 60 ppb O3 (daytime average for mid level O3 treatment), and three chambers were not fumigated with O3 and had an ambient concentration of approximately 15 ppb O3 (low-level ambient treatment, Table 1). The chambers for each treatment were chosen randomly (Appendix 1). Carbon dioxide and O3 concentrations simulated in this experiment represent current and predicted near-term future levels in the Phoenix metropolitan region (Idso, Idso, and Balling Jr. 2001; Wentz et al. 2002). Ozone and Carbon Dioxide monitoring: Ozone and CO2 concentrations were recorded in each chamber once every hour (taken as an average over 5 minute intervals) throughout the day. Air was pumped from an individual chamber through a scani-valve for six-minute intervals into a Dasibi O3 analyzer (Dasibi Environmental Corp, CA) and a CO2 IRGA (PP System EMG-4, MA) set-up independently of IRGAs inside each chamber. Both monitoring instruments were integrated with a chart recorder (Cole Palmer, Vernon Hills, IL) and PC data logger (Dataram 4, Thermo Instruments) for continuous data recording throughout the experiment. 211 N additions: Individual plants from each species were randomly divided into 3 groups (n = 15-17) to receive N treatments. N fertilizer was applied as ammonium nitrate (NH4- NO3) in de-ionized water solution made at concentrations equivalent to 0 (deionized -1 -1 water only), +4, and +8 kg N ha yr . NH4-NO3 solution was applied in four applications approximately every 10 – 12 days starting on day 1 of the experiment. Water trays were placed under the plants to capture any dripping; water trays were specific only to plants in the same N treatment. N treatments represented the range of ambient N deposition across the Phoenix urban-rural gradient (Lohse et al. 2008). Response variables I sampled plants from each treatment approximately every 7-10 days for a total of 5 sampling periods. However, the majority of remaining P. recurvata individuals in all treatments had senesced around the 4th sampling, and thus only S. arabicus was sampled five times in each treatment. At each “harvest,” I sampled 2-3 individual plants per species from each treatment (e.g., 2 species x 2 individuals x 27 treatments = 108 plants sampled per harvest). At each sampling, it took approximately 3-4 days to process and analyze all individuals; plants remained in the treatment chambers until the day they were to be sampled. Individuals were analyzed as pseudo-replicates and then averaged by treatment and sampling date to test the effects of CO2, O3, and N individually and in combination on each species’ dry mass, growth rate, and physiological characteristics. Physiological Fluorescence measurements: Chlorophyll fluorescence measurements were taken to quantify physiological stress on the Photosystem II (PS II) apparatus 212 (Maxwell and Johnson 2000). Each individual with minimal disturbance (plant intact in cone-tainer) was moved from the treatment chamber and placed in a dark box for at least 30 minutes to dark-adapt the plants for chlorophyll fluorescence measurements. Fluorescence was measured using a portable modulated chlorophyll fluorometer (PAM- 2000, Heinz Walz, Germany) following procedures described by Gremer and colleagues (2012) for small desert annual herbaceous species. Specifically, following the dark adaptation period, I used a non-saturating light pulse to measure initial fluorescence (Fo) and then a saturating light to estimate maximum fluorescence (Fm). Next, I applied an actinic light for 120 seconds before a second saturating pulse. Fluorescence level was measured before and after this second pulse to yield F's and F'm, respectively. The response to far-red light was then measured (F'o). Based on each response, I calculated the variable fluorescence (Fv = Fm – Fo), the maximum quantum yield of PSII (Fv/Fm) and the effective quantum yield of PS II (hereafter “yield”; yield = F'm-F's/F'm; Maxwell and Johnson 2000). Photosynthetic yield is a measure broadly describing photosynthetic stress by measuring light driven electron transport and CO2 assimilation, where lower yield values indicate a more stressed PS II (Henriques 2009). Leaf Area and Dry Mass: After fluorometric measurements were complete, plants were cut at the soil surface to separate below and aboveground biomass. Aboveground parts were then separated using forceps into leaf, stem, and reproductive parts, and belowground root mass was cleaned from potting soil. Reproductive biomass included any visible flower, bud or nutlet. Leaf area was measured using a portable leaf area meter (LI-COR 3000C, LI-COR Environmental, Nebraska). Leaves were placed flat without overlap between two clear plastic sheathes that passed through the LI-3000C 213 sensor head. Leaf area was measured three times for each sample and averaged for final area. For samples in which the leaves were too big to measure without overlap, the sample was divided into sub-samples, each measured as stated above, and the final averages were added together for the total leaf area for that individual. Finally, biomass was dried at 60oC for 48 hours and then weighed. Data analyses I analyzed treatment effects using linear mixed models in order to determine which factors (CO2, O3, and N) best predicted response variables for each species (lme4 package in R, R Core Team, 2014). The fixed effects included CO2, O3 and N treatments as single, additive (e.g., CO2 + O3) and interactive factors (e.g., CO2 * O3). Chamber was included in each model as a random effect in order to account for potential unmeasured differences among chambers. In order to test my hypotheses regarding the sensitivity of plants to single and co-occurring pollutants, I separately tested models with elevated CO2, O3, and N treatments as single fixed factors, combined additive terms or interactions terms (Table 2). I ran separate models for each species and dependent variable, including growth rate (aboveground biomass day-1) and physiological effective quantum yield of PS II (yield is unitless). Growth rate data were log transformed to meet basic assumptions of the linear mixed models. The growth rate did not differ between the final sampling periods (sampling period 4 and 5 for S. arabicus (F(1,167) = 1.5, p = 0.2) and sampling period 3 and 4 for P. recurvata (F(1,134) = 0.2, p = 0.7)), and so data were combined within each species for the final sampling periods in order to increase sample size for peak growth rate models. The 214 yield data were not aggregated. S. arabicus yield was analyzed from the fifth sampling, and P. recurvata yield was analyzed for the third sampling as this is before P. recurvata individuals began to senesce so the effects across all treatment combinations could be analyzed. To examine the importance of individual and co-occurring factors, I compared the goodness of fit of candidate mixed model results using Akaike’s information criterion adjusted for finite samples (AICc, Table 2). The AICc accounts for small sample sizes and lower AICc indicate a better goodness of fit (Burnham and Anderson 2002). The difference between AICc scores (Δ AICc = Model AICc – Best fit (lowest AICc) model AICc) was calculated to assist in choosing the best model. Models with Δ AICc less than 3 are more likely to be an alternative choice for best model compared to models with Δ AICc greater than 3 (Burnham and Anderson 2002). Candidate best models were nested and compared via likelihood ratio comparison tests in order to determine which individual, or set of additive or interaction factors are most important for predicting the response variable. After the best fit model was chosen, I compared each treatment to the control in each model with Dunnett’s (Dunnett 1955) many-to-one comparison test with a Bonferroni correction to account for potential inflated error from multiple comparisons. Finally, I tested if the global change factors have an overall additive, synergistic, or antagonistic effect. I calculated the predicted net effect (relative percent difference from the control) as the sum of individual treatments of elevated CO2, O3, or N. I compared the predicted net effect to the corresponding observed effects from empirical treatment combinations. For example, the predicted net growth rate of S. arabicus grown in combined elevated CO2 and O3 is 33.5% (33.5% is the sum of +44.3% net increased 215 growth when grown in 700 ppm CO2 and -10.9% change when grown in 100 ppb O3). The observed net growth in the combined treatment of CO2 and O3 is 25.5% (difference between predicted and observed = 8%). When the predicted change is equal (or comparable (<10% difference) to the observed change, this is an overall additive effect of the treatments. If the predicted change is more than 10% lower than observed change, the net response is synergistic (greater than the sum of individual effects), and if the predicted change is more than 10% greater than the observed response, the net response is antagonistic (less than the sum of individual effects). RESULTS At peak growth, the best fit models to explain growth rate of the non-native graminoid (S. arabicus) and native forb (P. recurvata) differed. S. arabicus’ growth rate and yield were best predicted by models including the interaction between CO2 and O3 (Table 2). P. recurvata’s growth rate was best predicted by O3 alone while a model including both CO2 and O3 was the best fit to describe physiological yield. Overall, elevated N or the interaction of N with CO2 or O3 was a minimally important predictor for the growth rate or yield of either species in this experiment (Table 2). Non-native species responses under single and combined factors The model including the interaction between elevated CO2 and O3 was the best predictive model of S. arabicus growth rate (Table 2). The best-fit model CO2 * O3 was determined by AICc value and likelihood ratio comparison between CO2 + O3 and CO2 * 216 2 O3 (χ = 9.1, p = 0.06). Nitrogen was not a significant factor in the growth rate model for S. arabicus (Table 2). As single factors, elevated CO2 increased growth rate and O3 exposure decreased the growth rate of S. arabicus (Figure 2, top panel). In combination, elevated CO2 mitigated the negative effects from exposure to elevated O3 (Figure 2, top panel). S. arabicus growth rate increased 34 and 44% in the elevated 550 ppm and 700 ppm CO2 treatments, respectively, compared to growth rate in ambient 400 ppm CO2 (averaged across all levels of O3 exposure, main effect of CO2 not shown). In contrast, growth rate decreased 5 and 11% in 60 ppb and 100 ppb O3 treatments compared to ambient 15 ppb O3 exposure (averaged across all levels of CO2 exposure, main effect of O3 not shown). At ambient O3, CO2 increased S. arabicus growth rate by an average 18% (9.7 +/- 0.6 mg day-1) and 36% (11.1 +/- 0.5 mg day-1) in 550 ppm and 700 ppm -1 CO2 treatments, respectively, compared to the control (8.2 +/- 0.3 mg day ; Figure 2 top panel). At ambient CO2, O3 as a single factor decreased growth rate by 9% (7.5 +/- 0.3 mg day-1) and 23% (6.3 +/- 0.2 mg day-1; Figure 2 top panel) under in 60 ppb and 100 ppb O3 exposure. However, at elevated CO2 concentrations, O3 damage was mitigated and there was a net positive increase in S. arabicus’ growth rate compared to the control (Dunnett’s test, p < 0.05 all high CO2 treatments, Figure 2 top panel). Similar to growth rate, the model accounting for the interaction between CO2 and O3 was the best predictive model of S. arabicus photosynthetic yield (i.e. model with lowest AICc score and no additional models with Δ AICc < 3; Table 2). Nitrogen was not significantly related to photosynthetic yield in S. arabicus (Table 2). In ambient CO2, photosynthetic yield was 54% (0.17 +/- 0.05) and 57% (0.16 +/- 0.04) lower than the control in the 60 ppb and 100 ppb O3 treatments, respectively (Dunnett’s test, p < 0.05, 217 Figure 2, bottom panel). A similar pattern followed for O3 exposure at 550 ppm CO2. In contrast, in high CO2 (700 ppm) exposure, photosynthetic yield did not differ from the control at any level of O3 exposure (Dunnett’s test, p > 0.1 all high CO2 treatments, Figure 2 bottom panel). The non-significant result, in comparison to the significant negative effects of O3 alone, indicates the negative impacts of O3 observed at lower CO2 concentrations were mitigated under elevated CO2 Native species affected most by O3 Contrary to S. arabicus, P. recurvata’s growth rate was best fit by a single factor model (Table 2, best fit model O3), and growth rate was not dependent on the interaction between CO2 and O3. Nitrogen was not significantly related to P. recurvata growth (Table 2). While the CO2 + O3 additive model and O3 single factor growth rate models had similar low AICc scores (Table 2), adding CO2 to the model did not significantly increase the model’s predictive power of P. recurvata’s growth rate (likelihood ratio comparison, χ2 = 2.5, p = 0.3). The growth rate of P. recurvata was negatively affected by exposure to elevated O3 concentrations. At ambient CO2, P. recurvata’s growth rate -1 decreased only 3% (4.7 +/- 0.6 mg day ) in the 60 ppb O3 treatment compared to the -1 -1 control (4.7 + /- 0.7 mg day ), but decreased 57% (2.0 +/- 0.5 mg day ) under high O3 (100 ppb) compared to the control (4.8 +/- 0.7 mg day-1, Figure 3, top panel). While exposure to elevated CO2 increased growth rate by 10% and 30% in 550 and 700 ppm CO2, respectively, the increased growth rate was not significantly different from the control due to variability among O3 treatments, variability of individual responses within treatments, and low sample size (Figure 3, top panel). 218 For the photosynthetic yield of P. recurvata, multiple mixed models were equally plausible best-fit candidates based on low AICc scores (Table 2, Δ AICc < 3). The model including CO2 and O3 without the interaction term (CO2 + O3) provided significantly better fit and lower residual error than the model with interaction term or O3 alone (CO2 + 2 O3, likelihood ratio comparison test χ = 15.1, p < 0.001). Elevated CO2 increased photosynthetic yield 34% at ambient O3 and 30% at mid-levels of O3, both of which are significantly greater than the control treatment (Dunnett’s test, p < 0.005, Figure 3, bottom panel). At ambient CO2, yield was reduced under the high O3 by 25% compared to the control, but was not significantly different than the control (Dunnett’s test, p = 0.2). Like S. arabicus, N was not significantly related to P. recurvata photosynthetic yield (Table 2). Additive and non-additive effects of combined factors I compared the percent change in growth rate (relative to the control) between species -1 -1 in the single factor high treatments (700 ppm CO2, 100 ppb O3, and 8 kg N ha yr ) and the combinations of these high treatments (Figure 4). Overall, non-native S. arabicus had net positive change in most treatments with only slight declines (-11%) in growth rate in high O3 treatments (100 ppb; Figure 4). In contrast, native P. recurvata was more sensitive to elevated O3. For example, the growth rate of P. recurvata declined 49% in high O3 exposure compared to the control (Figure 4). N only had a slight positive effect on S. arabicus (+5%) and no effect (-0.2%) on the growth rate of P. recurvata (Figure 4). Combined elevated CO2 and N (CO2 * N, Figure 4) increased growth rate 57 and 49% compared to the control for S. arabicus and P. recurvata, respectively. In contrast, 219 combined elevated O3 and N decreased growth rate 6% and 58% for S. arabicus and P. recurvata, respectively. N appeared to have a slight positive impact on the growth of S. arabicus where percent change in the combined O3 and N was lower (less negative) compared to that when exposed to O3 alone. On the other hand, the response of P. recurvata was a greater decline (more negative) in growth rate in the combined treatment compared to O3 alone. The greatest difference between species was in their response to combined CO2 and O3. The growth rate of P. recurvata decreased 30% from the control in combined CO2 and O3, while the growth rate of S. arabicus increased 25% despite co- occurring exposure to 100 ppb O3 (Figure 4). Finally, when exposed to all three pollutants in combination (CO2 x O3 x N), there was a net positive increase in growth rate by 49 and 11% for S. arabicus and P. recurvata, respectively. In order to further investigate if the combined factors had an overall additive, synergistic or antagonistic effect, I compared the predicted responses (summed effects -1 from individual treatments in combined high [700ppm CO2, 100ppb O3, and 8 kgN ha -1 -1 -1 yr ] and middle [550ppm CO2, 60ppb O3, and 4 kgN ha yr ] treatments) to the observed responses in corresponding experimental treatment combinations. Overall, there was net positive growth in treatment combinations of CO2 x N (Figure 4; triangles Figure 5) and CO2 x O3 x N for both species (Figure 4; diamonds, Figure 5), and net negative (less than the control) growth in combined O3 and N for both species (Figure 4; circles, Figure 5). Contrary to my expectations, the majority of S. arabicus’ growth rate responses to co- occurring pollutants were additive (i.e. predicted response approximately equals observed along the 1:1 line, Figure 5), while P. recurvata responses were more mixed. 220 For S. arabicus (dark grey and black symbols Figure 5), the observed growth effects were additive for all the two-way treatment combinations (CO2 x O3, CO2 x N, and O3 x N). However, there was a synergistic effect from all three pollutants combined (CO2 x O3 x N, diamonds Figure 5). The predicted net increase in S. arabicus’ growth rate in -1 -1 high treatments (700 ppm CO2 x 100 ppb O3 x 8 kg N ha yr N) was 38% compared to the observed 49% actual net increase in growth compared to the control (11% difference indicating a net synergistic effect, black diamond Figure 5). In the middle level treatments, predicted net increase in growth rate was 28% compared to 40% actual observed net increase (12% difference). The net effect of all three pollutants combined was greater than in two-way combinations of CO2 and O3 (-9% difference, square Figure 5). Thus, removing the effect of N appears to change the net response, though N was not a significant factor in the linear mixed models (Figure 5). Similar to S. arabicus, P. recurvata’s growth rate responses (open and light grey symbols, Figure 5) were additive in two-way combinations of CO2 and O3 (squares, Figure 5) and synergistic in response to all three pollutants combined (CO2 x O3 x N, diamonds, Figure 5). P. recurvata’s growth rate was predicted to decrease 23% when exposed concurrently to all three pollutants, however the observed response was a net increase of 11% compared to the control (34% difference indicating synergistic response, grey diamond Figure 5). There was also a synergistic response in the middle level treatments where the growth rate was predicted to increase 7%, but actually increased 30% (23% difference indicating synergistic response). I also observed net synergistic responses of P. recurvata grown in two-way combinations of elevated CO2 and N (triangles, Figure 5). In contrast, the net effect of O3 and N tended to be antagonistic 221 where there was a predicted decrease of 49% in the high treatments and 9% in the middle level treatments, but observed growth rate decreased 58% and 19%, respectively, a greater decrease than expected (circles, Figure 5). P. recurvata’s net negative growth as a result of O3 exposure (Figure 5 circles and squares) highlights the significance of O3 as a main influence on the native species’ growth (Figures 3 and 4). DISCUSSION I investigated the ecological impact of multiple ecologically relevant atmospheric compounds (CO2, N, and O3) on the growth and physiological yield of the dominant native and non-native Sonoran Desert herbaceous species. Overall, I found the two most frequently encountered species in this system, which differ in functional type and physiological tradeoffs, also differed in net responses and sensitivities to co-occurring pollutants. The non-native graminoid S. arabicus, characterized by high relative growth rate (RGR) and low water use efficiency (WUE), was most strongly influenced by elevated CO2. Elevated CO2 led to net positive growth of S. arabicus, despite the physiological stress of O3 on growth and photochemistry (Figure 2). In contrast, the native forb P. recurvata, characterized by high WUE and low RGR, showed positive physiological response to elevated CO2, but no accelerated growth (Figure 3). P. recurvata grew more slowly than controls under elevated O3 (Figure 3). Contrary to my expectations, elevated N had a minimal and non-significant impact on growth or physiological functioning for S. arabicus or P. recurvata (Table 2, Figure 4). Finally, I found that growth rate responses to combinations of two factors (O3 x N, CO2 x N, and 222 CO2 x O3) were largely additive (sum of individual responses) for the non-native species, while the native species responses were more frequently synergistic (amplifying or greater than the sum of individual effects, Figure 5). Overall, net responses to three-way combinations of co-occurring CO2, O3 and N were synergistic for both species, though the synergistic effect was more strongly apparent for the native forb (P. recurvata) than the non-native graminoid (S. arabicus, Figure 5). Responses to co-occurring pollutants The growth rate and physiological yield of both native and non-native species were affected differently by CO2 and O3 individually and in combination. At current global CO2 concentrations (i.e. concentrations comparable to ambient CO2 levels in this experiment), exposure to high O3 concentrations (100 ppb) reduced growth rate by 23% and 57% for the non-native and native species, respectively (Figure 2 & 3). The species differed, however, in their response to elevated CO2. Elevated CO2 mitigated the negative impacts of O3 exposure for the non-native. S. arabicus had net positive growth (greater than the control) when grown in 700 ppm CO2, regardless of co-occurring exposure to 60 or 100 ppb O3 concentrations (Figure 2, Figure 5). In contrast, elevated CO2 concentrations neither increased the growth rate in the native forb nor as strongly mitigated the negative effects of elevated O3 as seen in S. arabicus (Figure 3). Desert annual species differ in functional traits along a gradient of physiological tradeoffs between WUE and RGR and these differences can affect growth and physiological responses to elevated pollutants. For example, P. recurvata is characterized by relatively low stomatal conductance, high WUE and low RGR, whereas 223 S. arabicus tends to have higher stomatal conductance, RGR and lower WUE (Huxman et al. 2008; Angert et al. 2007; Angert et al. 2009). As expected, the peak growth rate of S. arabicus (8.2 +/- 0.3 mg day-1) was higher than P. recurvata in ambient conditions (4.8 +/- 0.7 mg day-1, Figure 2 & 3). I also expected S. arabicus to be more sensitive to exposure to elevated atmospheric pollutants, and O3 in particular, as a result of higher relative stomatal conductance. The oxidative stress of O3 damage is strongly correlated with stomatal conductance and directly impacts the photosynthetic mechanisms, Rubisco, and chlorophyll content (Reich and Amundson 1985). I measured photosynthetic quantum yield with chlorophyll fluorescence as an indicator of photosynthetic stress from exposure to pollutants. As expected, S. arabicus had significant physiological stress in response to elevated O3, whereas P. recurvata did not exhibit lower yields or physiological stress from O3 exposure. S. arabicus’ yield was reduced up to 57% when exposed to elevated O3 at ambient (400 ppm) and enriched (550 ppm) CO2 (Figure 2). Yet, as with growth rate, photosynthetic stress of S. arabicus at high O3 was reduced when concurrently exposed to elevated CO2 (Figure 2). The net positive physiological response of S. arabicus in part explains the mechanism for net positive growth in elevated O3 and CO2. As CO2 concentrations rise and the photosynthetic need for CO2 uptake is reduced, stomatal conductance is also reduced, which in turn minimizes the flux of O3 into the leaf and mitigates O3 stress and damage (Cardoso‐Vilhena and Barnes 2001; Cardoso‐Vilhena et al. 2004). Despite the physiological stress from O3 exposure, the net positive growth rate highlights the plasticity of S. arabicus to respond and adapt to increasing environmental conditions and concentrations of co-occurring pollutants. 224 The physiological response of native P. recurvata was more variable. Elevated CO2 increased photosynthetic yield up to 34% in some treatments, yet elevated O3 did not significantly reduce yield of the native species (Figure 3). Characterized by relatively low stomatal conductance, physiological damage from O3 is minimal in P. recurvata indicating the P. recurvata may be more physiologically stress tolerant. However, despite the minimal physiological response to elevated O3, the growth rate of P. recurvata was significantly reduced when exposed to high (100 ppb) O3 concentrations as a result of observed physical damage and necrosis of leaf biomass. Overall, the physical damage and lack of interaction between CO2 and O3 suggests the non-native species has less plasticity to respond to changing environmental conditions. Contrary to my expectations, elevated N had minimal predictive power in explaining growth or physiological responses of either species (Table 2, Figure 4). I expected elevated N in combination with elevated CO2 would reduce co-limitations on growth and lead to synergistic effects on growth rate (Shen et al. 2008). For example, I saw additive growth rate responses when plants were exposed to co-occurring CO2 and O3, but in combinations including elevated N, the responses were synergistic (diamonds above the 1:1 line, Figure 5). The lack of power in the models including N (Table 2) to predict these synergistic relationships may be due, in part, to low sample size in the combined CO2, O3, N treatments. However, the limited N impact in this experiment was also consistent with a multi-year field study from the Sonoran Desert, in which desert herbaceous biomass did not differ between urban and outlying locations despite larger soil inorganic N pools and higher N deposition in the urban locations (Hall et al. 2011). In wet years, N was limiting to desert herbaceous plants, and plants responded under high 225 -1 -1 rates of N additions (60 kg N ha yr ) in a field manipulation experiment. Similarly, the desert herbaceous species in my study may not be responsive to relatively low N inputs (8 kg N ha-1 yr-1 applied in this study compared to 60 kg N ha-1 yr-1 in field manipulation) that are characteristic of the N deposited in the Phoenix region, even under well-watered conditions. Overall, however, background inorganic N in the homogenized soil mixture used as a growing medium likely explains the minimal effects of my experimental N treatments. High background levels in the soil (~45 ppm) most likely overshadowed the -1 -1 N I applied as ammonium nitrate in solution to simulate 4 and 8 kg N ha yr deposition (6 and 12 ppm respectively). Thus, I expect N may not have been limiting for the plants in any of the treatments. Predicted impacts of water co-limitation In arid ecosystems where water is the main limiting factor, aridland plants are expected to respond differently to environmental factors, such as elevated CO2, depending on water availability. I maintained well-watered plants and elevated soil moisture to reduce variability in the high number of treatment. Similar to findings in other arid studies, I found elevated CO2 in high water conditions increased non-native grass biomass more than the native species (S. D. Smith et al. 2000). However, I expect my data may best describe the “potential” growth rate when WUE is less essential to the survival in well-watered conditions, and the results may vary under more water-limited conditions. For example, in a long-term manipulative experiment in the Mojave Desert, herbaceous plants have increased production and reproductive allocation in response to elevated CO2 in wet years but not drought years, which can significantly influence the 226 long-term dominance of native and non-native species (S. D. Smith et al. 2000; Housman et al. 2006; S. D. Smith et al. 2014). In contrast, other studies report mixed effects of elevated CO2 on biomass during wet and dry years, where in drier years elevated CO2 in arid ecosystems results in greater water use efficiency, in turn leading to increased soil moisture and the potential for increased N mineralization and uptake (Morgan et al. 2004; Dijkstra et al. 2010). Thus, aridland C3 herbaceous production is expected to increase non-linearly in response to elevated CO2 and N (i.e., a synergistic effect) in wet years, while in dry years the effects would be antagonistic (Shen et al 2008). In addition to the effects of elevated CO2 concentrations, this study is novel in also testing the co-occurring impacts of other pollutants. I found the non-native species was particularly sensitive to elevated O3 exposure at 60 and 100 ppb O3 (Figure 2). Under more water-limited conditions, I would expect S. arabicus to be less sensitive to O3 with perhaps greater interactive effects in combinations of elevated CO2 and O3. Finally, herbaceous annual plants’ functional traits of WUE and RGR, which are consistent across years and experimental conditions (e.g. greenhouse vs field experiment), can be translated to a tradeoff in fitness under different environmental conditions, such as water availability (Kimball et al. 2012). For example, in wet years, high RGR species (e.g. S. arabicus) are expected to outperform species with lower RGR and higher WUE (e.g. P. recurvata) as a result of greater plasticity and an ability to exploit resources (Kimball et al. 2012). The findings in this study support this, as elevated CO2 concentrations led to an overall increased growth rate of S. arabicus despite damage from elevated O3. As in other studies, I expect that in drier conditions, the native forb may be favored due to its relatively high WUE and ability to tolerate physiological 227 stress (S. D. Smith et al. 2014). Yet, our findings highlight the native forb’s susceptibility to O3 damage to leaves, despite low physiological stress. These results also highlight the important role of water as a driving factor in arid ecosystem responses to multiple pollutants and the need to test co-occurring factors with water as an additional constraint. Synergistic responses to realistic co-occurring pollutants I expected co-occurring elevated pollutants to have interactive effects leading to overall net synergistic or antagonistic growth responses in each species. With the full factorial dose response design, I was able to ask if net growth rates were simply additive and could have been predicted by studying each factor individually, or if responses were synergistic or antagonistic requiring a more complex model to predict the results. Overall, I found co-occurring elevated CO2 and O3 had an interactive effect on the photosynthetic yield and growth rate of non-native S. arabicus, such that CO2 largely mitigated the negative effects of elevated O3 (Table 2, Figure 3). On the other hand, net negative growth of P. recurvata was observed in the same treatment (Figures 2 and 3). These results suggest that CO2 was a stronger driver of S. arabicus growth, while O3 was the stronger control on P. recurvata growth. Nevertheless, despite the significant interaction between CO2 and O3 for S. arabicus, the net growth rate of both species was additive in combinations of CO2 and O3 (i.e., predicted and observed effects were equal; squares, Figure 5). On the other hand, both species had net synergistic growth rate responses to combined CO2, O3, and N (Figure 5). P. recurvata also had a synergistic growth response to elevated CO2 and N (Figure 5). Though N was not a significant 228 factor in the models, likely due to the background soil N used in this experiment, the non- additive results suggest possible synergistic responses of both species grown in the ecological airshed of cities, in which elevated CO2, O3, and N deposition co-occur. It is essential to identify when ecosystem response may be synergistic or antagonistic, as possible non-additive responses increase the complexity of systems models needed to accurately predict future ecological outcomes. Additional multi-factor experiments and models are needed to assess the potential synergistic responses to realistic combinations of elevated CO2, O3, and N in combination with species interactions and other co- occurring global changes such as altered temperature and precipitation patterns (Bytnerowicz et al 2013, Templer 2013). Implications of urban air quality for primary producers and ecosystems Current and future levels of multiple pollutants may lead to long-term changes in plant community composition, yet species and community responses to co-occurring future levels of CO2, O3, and N are unknown. For example, elevated N deposition is expected to reduce species diversity even at chronic, low levels of N inputs to an ecosystem (Clark and Tilman 2008; Payne et al. 2013). On the other hand, elevated CO2 has mixed effects on changing community composition (Potvin and Vasseur 1997; Zavaleta et al. 2003). Yet, in combination, elevated CO2 has been found to mitigate the negative effects of N deposition on species loss (Reich 2009). These studies, however, do not account for the co-occurring exposure to elevated CO2, N, and O3 in the “ecological airshed” in native ecosystems surrounding cities (Cook et al in prep). 229 Though I did not test community-level responses to co-occurring CO2, N, and O3, the net responses of the dominant species of the Sonoran Desert differed, such that the non- native species will be favored in predicted future environmental conditions. In combinations of predicted current and future CO2 and O3 concentrations, the non-native species had overall net positive growth despite co-occurring O3 and negative physiological responses to O3. This suggests elevated CO2 had a protective effect for the non-native species under high stress. While the native species responded similarly with reduced growth in elevated O3, native P. recurvata was less physiologically sensitive to O3. Overall, elevated CO2 did not have the same mitigating effects on O3 responses for the native species. The different net responses could lead to the non-native species exploitation of resources and outcompeting the native species within the urban “ecological airshed.” The eco-physiological tradeoffs between WUE and RGR in part explain the current co-existence of Sonoran Desert species, as well as the variability in how species respond to current and predicted future urban pollutants (Angert et al. 2009; Kimball et al. 2012). However, additional changing environmental patterns, such as longer periods of drought and increased temperature, will interact with eco-physiological traits and elevated pollutant levels with potential cascading impacts on the co-existence, competition, and dominance of herbaceous species (Kimball et al 2012). More research is needed to specifically link the eco-physiological responses of species to community level changes in order to better predict changing ecosystem structure (Suding et al. 2008; Crous et al. 2010). Changes to the community structure can have important ecosystem consequences: for example, altered community composition of arid grasslands may lead 230 to increased fire frequency and altered ecosystem services, such as the ability to store carbon and N (S. D. Smith et al. 2000; Hall et al. 2011). CONCLUSION Decades of research have shown that plant processes are sensitive to altered climate and elevated concentrations of CO2, N, or O3 alone, as they function as both resources and stressors for primary producers (Long et al. 2004; Karnosky et al. 2007; Bobbink et al. 2010). While CO2, O3 and N are expected to be elevated concurrently as a result of human activities, particularly in and around cities, no research to date has examined their co-occurring impacts at concentrations that reflect current and predicted future levels along urban-rural gradients. These results show that native ecosystems are vulnerable to current and future air pollution over the long-term. The findings in this study highlight that current and predicted future air quality conditions in and around urban regions can have important impacts, including interactive and synergistic effects, on individual species growth and physiological functioning with potential cascading impacts on the community composition of protected ecosystems. 231 ACKNOWLEDGEMENTS This work was completed in collaboration with Sharon Hall (Arizona State University) and Pamela Padgett (USFS, Riverside, CA). Thank you to David Jones and Tait Rounsville (USFS, Riverside CA), Elizabeth Hessom, Darrel Jenerette, and Lou Santiago (UC Riverside), Dan Childers and members of the Hall lab (ASU) for lab help, use of facilities and equipment, as well as ongoing discussions about this research. This work was work supported by the National Science Foundation under grant no. DEB-0423704 and BCS-1026865 CAP LTER, grant no. 0504248, Integrative Graduate Education Research and Training (IGERT) in Urban Ecology, and the Arizona State University School of Life Sciences Lisa Dent Memorial Scholarship fund. 232 REFERENCES Aber, John D., W. McDowell, K. Nadelhoffer, A. Magill, G. Berntson, M. Kamakea, S. McNulty, W. Currie, L. Rustad, and Ivan Fernandez. 1998. “Nitrogen Saturation in Temperate Forests: Hypotheses Revisited.” Bioscience 48 (11): 921–34. Aber, John D., Scott V. Ollinger, Charles T. Driscoll, Gene E. Likens, Richard T. Holmes, Rita J. Freuder, and Christine L. Goodale. 2002. “Inorganic Nitrogen Losses from a Forested Ecosystem in Responseto Physical, Chemical, Biotic,and Climatic Perturbations.” Ecosystems 5 (7): 0648–58. doi:10.1007/s10021-002- 0203-8. Ainsworth, Elizabeth A., Craig R. Yendrek, Stephen Sitch, William J. Collins, and Lisa D. Emberson. 2012. “The Effects of Tropospheric Ozone on Net Primary Productivity and Implications for Climate Change*.” Annual Review of Plant Biology 63 (1): 637–61. doi:10.1146/annurev-arplant-042110-103829. Akimoto, Hajime. 2003. “Global Air Quality and Pollution.” Science 302 (5651): 1716– 19. doi:10.1126/science.1092666. Angert, A. L., T. E. Huxman, G. A. Barron-Gafford, K. L. Gerst, and D. L. Venable. 2007. “Linking Growth Strategies to Long-Term Population Dynamics in a Guild of Desert Annuals.” Journal of Ecology 95 (2): 321–31. doi:10.1111/j.1365- 2745.2006.01203.x. Angert, A. L., Travis E. Huxman, Peter Chesson, and D. Lawrence Venable. 2009. “Functional Tradeoffs Determine Species Coexistence via the Storage Effect.” Proceedings of the National Academy of Sciences 106 (28): 11641–45. doi:10.1073/pnas.0904512106. Báez, S., J. Fargione, D. I. Moore, S. L. Collins, and J. R. Gosz. 2007. “Atmospheric Nitrogen Deposition in the Northern Chihuahuan Desert: Temporal Trends and Potential Consequences.” Journal of Arid Environments 68 (4): 640–51. doi:10.1016/j.jaridenv.2006.06.011. Bassin, Seraina, Roland A. Werner, Karin Sörgel, Matthias Volk, Nina Buchmann, and Jürg Fuhrer. 2009. “Effects of Combined Ozone and Nitrogen Deposition on the in Situ Properties of Eleven Key Plant Species of a Subalpine Pasture.” Oecologia 158 (4): 747–56. doi:10.1007/s00442-008-1191-y. Bergeron, Onil, and Ian B. Strachan. 2011. “CO2 Sources and Sinks in Urban and Suburban Areas of a Northern Mid-Latitude City.” Atmospheric Environment 45 (8): 1564–73. doi:10.1016/j.atmosenv.2010.12.043. 233 Bobbink, Roland, K. Hicks, J. Galloway, T. Spranger, R. Alkemade, M. Ashmore, M. Bustamante, S. Cinderby, E. Davidson, and F. Dentener. 2010. “Global Assessment of Nitrogen Deposition Effects on Terrestrial Plant Diversity: A Synthesis.” Ecological Applications 20 (1): 30–59. Brooks, Matthew L. 2003. “Effects of Increased Soil Nitrogen on the Dominance of Alien Annual Plants in the Mojave Desert.” Journal of Applied Ecology 40 (2): 344–53. doi:10.1046/j.1365-2664.2003.00789.x. Bytnerowicz, A., D. M. Olszyk, C. A. Fox, P. J. Dawson, G. Kats, C. L. Morrison, and J. Wolf. 1988. “Responses of Desert Annual Plants to Ozone and Water Stress in an in Situ Experiment.” JAPCA 38 (9): 1145–51. doi:10.1080/08940630.1988.10466463. Cardoso‐Vilhena, João, Luis Balaguer, Derek Eamus, John Ollerenshaw, and Jeremy Barnes. 2004. “Mechanisms Underlying the Amelioration of O3‐induced Damage by Elevated Atmospheric Concentrations of CO2.” Journal of Experimental Botany 55 (397): 771–81. doi:10.1093/jxb/erh080. Cardoso‐Vilhena, João, and Jeremy Barnes. 2001. “Does Nitrogen Supply Affect the Response of Wheat (Triticum Aestivum Cv. Hanno) to the Combination of Elevated CO2 and O3?” Journal of Experimental Botany 52 (362): 1901–11. doi:10.1093/jexbot/52.362.1901. Chapin, F. Stuart, III. 1980. “The Mineral Nutrition of Wild Plants.” Annual Review of Ecology and Systematics 11 (January): 233–60. ———. 1991. “Integrated Responses of Plants to Stress.” BioScience 41 (1): 29–36. doi:10.2307/1311538. Clark, Christopher M., and David Tilman. 2008. “Loss of Plant Species after Chronic Low-Level Nitrogen Deposition to Prairie Grasslands.” Nature 451 (7179): 712– 15. doi:10.1038/nature06503. Crous, Kristine Y., Peter B. Reich, Mark D. Hunter, and David S. Ellsworth. 2010. “Maintenance of Leaf N Controls the Photosynthetic CO2 Response of Grassland Species Exposed to 9 Years of Free-Air CO2 Enrichment.” Global Change Biology 16 (7): 2076–88. doi:10.1111/j.1365-2486.2009.02058.x. Dentener, F., D. Stevenson, K. Ellingsen, T. van Noije, M. Schultz, M. Amann, C. Atherton, et al. 2006. “The Global Atmospheric Environment for the Next Generation.” Environmental Science & Technology 40 (11): 3586–94. doi:10.1021/es0523845. 234 Dijkstra, Feike A., Dana Blumenthal, Jack A. Morgan, Daniel R. LeCain, and Ronald F. Follett. 2010. “Elevated CO2 Effects on Semi-Arid Grassland Plants in Relation to Water Availability and Competition.” Functional Ecology 24 (5): 1152–61. doi:10.1111/j.1365-2435.2010.01717.x. Dukes, Jeffrey S, Nona R Chiariello, Elsa E Cleland, Lisa A Moore, M. Rebecca Shaw, Susan Thayer, Todd Tobeck, Harold A Mooney, and Christopher B Field. 2005. “Responses of Grassland Production to Single and Multiple Global Environmental Changes.” PLoS Biol 3 (10): e319. doi:10.1371/journal.pbio.0030319. Elser, James J., Matthew E.S. Bracken, Elsa E. Cleland, Daniel S. Gruner, W. Stanley Harpole, Helmut Hillebrand, Jacqueline T. Ngai, Eric W. Seabloom, Jonathan B. Shurin, and Jennifer E. Smith. 2007. “Global Analysis of Nitrogen and Phosphorus Limitation of Primary Producers in Freshwater, Marine and Terrestrial Ecosystems.” Ecology Letters 10 (12): 1135–42. doi:10.1111/j.1461- 0248.2007.01113.x. Fangmeier, A, L De Temmerman, C Black, K Persson, and V Vorne. 2002. “Effects of Elevated CO2 And/or Ozone on Nutrient Concentrations and Nutrient Uptake of Potatoes.” European Journal of Agronomy 17 (4): 353–68. doi:10.1016/S1161- 0301(02)00071-0. Felzer, Benjamin S., Timothy W. Cronin, Jerry M. Melillo, David W. Kicklighter, and C. Adam Schlosser. 2009. “Importance of Carbon-Nitrogen Interactions and Ozone on Ecosystem Hydrology during the 21st Century.” Journal of Geophysical Research: Biogeosciences 114 (G1): G01020. doi:10.1029/2008JG000826. Fenn, Mark E., E. B. Allen, S. B. Weiss, S. Jovan, L. H. Geiser, G. S. Tonnesen, R. F. Johnson, et al. 2010. “Nitrogen Critical Loads and Management Alternatives for N-Impacted Ecosystems in California.” Journal of Environmental Management 91 (12): 2404–23. doi:10.1016/j.jenvman.2010.07.034. George, K., L. H. Ziska, J. A. Bunce, and B. Quebedeaux. 2007. “Elevated Atmospheric CO2 Concentration and Temperature across an Urban–rural Transect.” Atmospheric Environment 41 (35): 7654–65. doi:10.1016/j.atmosenv.2007.08.018. Gregg, Jillian W., Clive G. Jones, and Todd E. Dawson. 2003. “Urbanization Effects on Tree Growth in the Vicinity of New York City.” Nature 424 (6945): 183–87. doi:10.1038/nature01728. Gremer, Jennifer R., Sarah Kimball, A. L. Angert, D. Lawrence Venable, and Travis E. Huxman. 2012. “Variation in Photosynthetic Response to Temperature in a Guild of Winter Annual Plants.” Ecology 93 (12): 2693–2704. doi:10.1890/12-0006.1. 235 Grimm, Nancy B., Stanley H. Faeth, Nancy E. Golubiewski, Charles L. Redman, Jianguo Wu, Xuemei Bai, and John M. Briggs. 2008. “Global Change and the Ecology of Cities.” Science 319 (5864): 756–60. doi:10.1126/science.1150195. Groffman, Peter M., Jill S. Baron, Tamara Blett, Arthur J. Gold, Iris Goodman, Lance H. Gunderson, Barbara M. Levinson, et al. 2006. “Ecological Thresholds: The Key to Successful Environmental Management or an Important Concept with No Practical Application?” Ecosystems 9 (1): 1–13. doi:10.1007/s10021-003-0142-z. Hall, Sharon J., Ryan A. Sponseller, Nancy B. Grimm, David Huber, Jason P. Kaye, Christopher Clark, and Scott L. Collins. 2011. “Ecosystem Response to Nutrient Enrichment across an Urban Airshed in the Sonoran Desert.” Ecological Applications 21 (3): 640–60. doi:10.1890/10-0758.1. Heagle, A. S., J. E. Miller, F. L. Booker, and W. A. Pursley. 1999. “Ozone Stress, Carbon Dioxide Enrichment, and Nitrogen Fertility Interactions in Cotton.” Crop Science 39 (3): 731. doi:10.2135/cropsci1999.0011183X003900030021x. Henriques, Fernando S. 2009. “Leaf Chlorophyll Fluorescence: Background and Fundamentals for Plant Biologists.” The Botanical Review 75 (3): 249–70. doi:10.1007/s12229-009-9035-y. Henry, Hugh A. L., Nona R. Chiariello, Peter M. Vitousek, Harold A. Mooney, and Christopher B. Field. 2006. “Interactive Effects of Fire, Elevated Carbon Dioxide, Nitrogen Deposition, and Precipitation on a California Annual Grassland.” Ecosystems 9 (7): 1066–75. doi:10.1007/s10021-005-0077-7. Horowitz, Larry W. 2006. “Past, Present, and Future Concentrations of Tropospheric Ozone and Aerosols: Methodology, Ozone Evaluation, and Sensitivity to Aerosol Wet Removal.” Journal of Geophysical Research: Atmospheres 111 (D22): n/a– n/a. doi:10.1029/2005JD006937. Housman, David C., Elke Naumburg, Travis E. Huxman, Therese N. Charlet, Robert S. Nowak, and Stanley D. Smith. 2006. “Increases in Desert Shrub Productivity under Elevated Carbon Dioxide Vary with Water Availability.” Ecosystems 9 (3): 374–85. doi:10.1007/s10021-005-0124-4. Huxman, Travis E., Greg Barron-Gafford, Katharine L. Gerst, Amy L. Angert, Anna P. Tyler, and D. Lawrence Venable. 2008. “Photosynthetic Resource Use Efficiency and Demographic Variability in Desert Winter Annual Plants.” Ecology 89 (6): 1554–63. doi:10.1890/06-2080.1. Idso, Craig D, Sherwood B Idso, and Robert C Balling Jr. 2001. “An Intensive Two- Week Study of an Urban CO2 Dome in Phoenix, Arizona, USA.” Atmospheric 236 Environment 35 (6): 995–1000. doi:10.1016/S1352-2310(00)00412-X. IPCC. 2014. “Intergovernmental Panel on Climate Change: Fifth Assessment Report.” http://www.ipcc.ch/. Karnosky, David F., John M. Skelly, Kevin E. Percy, and Art H. Chappelka. 2007. “Perspectives Regarding 50 Years of Research on Effects of Tropospheric Ozone Air Pollution on US Forests.” Environmental Pollution 147 (3): 489–506. doi:10.1016/j.envpol.2006.08.043. Kimball, Sarah, Jennifer R. Gremer, Amy L. Angert, Travis E. Huxman, and D. Lawrence Venable. 2012. “Fitness and Physiology in a Variable Environment.” Oecologia 169 (2): 319–29. doi:10.1007/s00442-011-2199-2. Krupa, S. V. 2003. “Effects of Atmospheric Ammonia (NH3) on Terrestrial Vegetation: A Review.” Environmental Pollution 124 (2): 179–221. doi:10.1016/S0269- 7491(02)00434-7. Langley, J. Adam, and J. Patrick Megonigal. 2010. “Ecosystem Response to Elevated CO2 Levels Limited by Nitrogen-Induced Plant Species Shift.” Nature 466 (7302): 96–99. doi:10.1038/nature09176. LeBauer, David S., and Kathleen K. Treseder. 2008. “NITROGEN LIMITATION OF NET PRIMARY PRODUCTIVITY IN TERRESTRIAL ECOSYSTEMS IS GLOBALLY DISTRIBUTED.” Ecology 89 (2): 371–79. doi:10.1890/06-2057.1. Lindroth, Richard L. 2010. “Impacts of Elevated Atmospheric CO2 and O3 on Forests: Phytochemistry, Trophic Interactions, and Ecosystem Dynamics.” Journal of Chemical Ecology 36 (1): 2–21. doi:10.1007/s10886-009-9731-4. Lohse, Kathleen A., Diane Hope, Ryan Sponseller, Jonathan O. Allen, and Nancy B. Grimm. 2008. “Atmospheric Deposition of Carbon and Nutrients across an Arid Metropolitan Area.” Science of The Total Environment 402 (1): 95–105. doi:10.1016/j.scitotenv.2008.04.044. Lombardozzi, D., J. P. Sparks, and G. Bonan. 2013. “Integrating O3 Influences on Terrestrial Processes: Photosynthetic and Stomatal Response Data Available for Regional and Global Modeling.” Biogeosciences Discussions 10 (4): 6973–7012. doi:10.5194/bgd-10-6973-2013. Long, Stephen P, Elizabeth A Ainsworth, Alistair Rogers, and Donald R Ort. 2004. “Rising Atmospheric Carbon Dioxide: Plants FACE the Future.” Annual Review of Plant Biology 55: 591–628. doi:10.1146/annurev.arplant.55.031903.141610. 237 Lovett, Gary M., Timothy H. Tear, David C. Evers, Stuart E.G. Findlay, B. Jack Cosby, Judy K. Dunscomb, Charles T. Driscoll, and Kathleen C. Weathers. 2009. “Effects of Air Pollution on Ecosystems and Biological Diversity in the Eastern United States.” Annals of the New York Academy of Sciences 1162 (1): 99–135. doi:10.1111/j.1749-6632.2009.04153.x. Luo, Yiqi, and James F. Reynolds. 1999. “VALIDITY OF EXTRAPOLATING FIELD CO2 EXPERIMENTS TO PREDICT CARBON SEQUESTRATION IN NATURAL ECOSYSTEMS.” Ecology 80 (5): 1568–83. doi:10.1890/0012- 9658(1999)080[1568:VOEFCE]2.0.CO;2. Maxwell, Kate, and Giles N. Johnson. 2000. “Chlorophyll Fluorescence—a Practical Guide.” Journal of Experimental Botany 51 (345): 659–68. doi:10.1093/jexbot/51.345.659. Melillo, Jm, Ad Mcguire, Dw Kicklighter, B. Moore, Cj Vorosmarty, and Al Schloss. 1993. “Global Climate-Change and Terrestrial Net Primary Production.” Nature 363 (6426): 234–40. doi:10.1038/363234a0. Monks, P. S., C. Granier, S. Fuzzi, A. Stohl, M. L. Williams, H. Akimoto, M. Amann, et al. 2009. “Atmospheric Composition Change – Global and Regional Air Quality.” Atmospheric Environment, ACCENT Synthesis, 43 (33): 5268–5350. doi:10.1016/j.atmosenv.2009.08.021. Morgan, J. A., D. E. Pataki, C. Körner, H. Clark, S. J. Del Grosso, J. M. Grünzweig, A. K. Knapp, et al. 2004. “Water Relations in Grassland and Desert Ecosystems Exposed to Elevated Atmospheric CO2.” Oecologia 140 (1): 11–25. doi:10.1007/s00442-004-1550-2. Ollinger, Scott V., John D. Aber, Peter B. Reich, and Rita J. Freuder. 2002. “Interactive Effects of Nitrogen Deposition, Tropospheric Ozone, Elevated CO2 and Land Use History on the Carbon Dynamics of Northern Hardwood Forests.” Global Change Biology 8 (6): 545–62. doi:10.1046/j.1365-2486.2002.00482.x. Padgett, Pamela E., Andrzej Bytnerowicz, Philip J. Dawson, George H. Riechers, and Dennis R. Fitz. 2004. “Design, Evaluation and Application of a Continuously Stirred Tank Reactor System for Use in Nitric Acid Air Pollutant Studies.” Water, Air, and Soil Pollution 151 (1-4): 35–51. doi:10.1023/B:WATE.0000009890.74470.fa. Pardo, Linda H., Mark E. Fenn, Christine L. Goodale, Linda H. Geiser, Charles T. Driscoll, Edith B. Allen, Jill S. Baron, et al. 2011. “Effects of Nitrogen Deposition and Empirical Nitrogen Critical Loads for Ecoregions of the United States.” Ecological Applications 21 (8): 3049–82. doi:10.1890/10-2341.1. 238 Payne, Richard J., Nancy B. Dise, Carly J. Stevens, David J. Gowing, Cecilia Duprè, Edu Dorland, Cassandre Gaudnik, et al. 2013. “Impact of Nitrogen Deposition at the Species Level.” Proceedings of the National Academy of Sciences 110 (3): 984– 87. doi:10.1073/pnas.1214299109. Potvin, Catherine, and Liette Vasseur. 1997. “LONG-TERM CO2 ENRICHMENT OF A PASTURE COMMUNITY: SPECIES RICHNESS, DOMINANCE, AND SUCCESSION.” Ecology 78 (3): 666–77. doi:10.1890/0012- 9658(1997)078[0666:LTCEOA]2.0.CO;2. Reich, Peter B. 2009. “Elevated CO2 Reduces Losses of Plant Diversity Caused by Nitrogen Deposition.” Science 326 (5958): 1399–1402. doi:10.1126/science.1178820. Reich, Peter B., and Robert G. Amundson. 1985. “Ambient Levels of Ozone Reduce Net Photosynthesis in Tree and Crop Species.” Science 230 (4725): 566–70. doi:10.1126/science.230.4725.566. Reich, Peter B., and Sarah E. Hobbie. 2013. “Decade-Long Soil Nitrogen Constraint on the CO2 Fertilization of Plant Biomass.” Nature Climate Change 3 (3): 278–82. doi:10.1038/nclimate1694. Reich, Peter B., Sarah E. Hobbie, Tali Lee, David S. Ellsworth, Jason B. West, David Tilman, Johannes M. H. Knops, Shahid Naeem, and Jared Trost. 2006. “Nitrogen Limitation Constrains Sustainability of Ecosystem Response to CO2.” Nature 440 (7086): 922–25. doi:10.1038/nature04486. Reid, Chantal D., and Edwin L. Fiscus. 2008. “Ozone and Density Affect the Response of Biomass and Seed Yield to Elevated CO2 in Rice.” Global Change Biology 14 (1): 60–76. doi:10.1111/j.1365-2486.2007.01472.x. Rustad, Lindsey E. 2008. “The Response of Terrestrial Ecosystems to Global Climate Change: Towards an Integrated Approach.” Science of The Total Environment, BIOGEOCHEMISTRY OF FORESTED ECOSYSTEM - Selected papers from BIOGEOMON, the 5th International Symposium on Ecosystem Behaviour, held at the University of California, Santa Cruz, on June 25–30, 2006, 404 (2–3): 222– 35. doi:10.1016/j.scitotenv.2008.04.050. Sala, O. E., F. Stuart Chapin, Iii, Juan J. Armesto, Eric Berlow, Janine Bloomfield, Rodolfo Dirzo, et al. 2000. “Global Biodiversity Scenarios for the Year 2100.” Science 287 (5459): 1770–74. doi:10.1126/science.287.5459.1770. Sala, O. E., and W. K. Lauenroth. 1982. “Small Rainfall Events: An Ecological Role in Semiarid Regions.” Oecologia 53 (3): 301–4. doi:10.1007/BF00389004. 239 Seto, Karen C., Burak Güneralp, and Lucy R. Hutyra. 2012. “Global Forecasts of Urban Expansion to 2030 and Direct Impacts on Biodiversity and Carbon Pools.” Proceedings of the National Academy of Sciences 109 (40): 16083–88. doi:10.1073/pnas.1211658109. Shaw, M. Rebecca, Erika S. Zavaleta, Nona R. Chiariello, Elsa E. Cleland, Harold A. Mooney, and Christopher B. Field. 2002. “Grassland Responses to Global Environmental Changes Suppressed by Elevated CO2.” Science 298 (5600): 1987–90. doi:10.1126/science.1075312. Shen, Weijun, Jianguo Wu, Nancy B. Grimm, and Diane Hope. 2008. “Effects of Urbanization-Induced Environmental Changes on Ecosystem Functioning in the Phoenix Metropolitan Region, USA.” Ecosystems 11 (1): 138–55. doi:10.1007/s10021-007-9085-0. Smith, Melinda D., Alan K. Knapp, and Scott L. Collins. 2009. “A Framework for Assessing Ecosystem Dynamics in Response to Chronic Resource Alterations Induced by Global Change.” Ecology 90 (12): 3279–89. doi:10.1890/08-1815.1. Smith, Stanley D., Therese N. Charlet, Stephen F. Zitzer, Scott R. Abella, Cheryl H. Vanier, and Travis E. Huxman. 2014. “Long-Term Response of a Mojave Desert Winter Annual Plant Community to a Whole-Ecosystem Atmospheric CO2 Manipulation (FACE).” Global Change Biology 20 (3): 879–92. doi:10.1111/gcb.12411. Smith, Stanley D., Travis E. Huxman, Stephen F. Zitzer, Therese N. Charlet, David C. Housman, James S. Coleman, Lynn K. Fenstermaker, Jeffrey R. Seemann, and Robert S. Nowak. 2000. “Elevated CO2 Increases Productivity and Invasive Species Success in an Arid Ecosystem.” Nature 408 (6808): 79–82. doi:10.1038/35040544. Suding, Katharine N., Sandra Lavorel, F. S. Chapin, Johannes H. C. Cornelissen, Sandra Díaz, Eric Garnier, Deborah Goldberg, David U. Hooper, Stephen T. Jackson, and Marie-Laure Navas. 2008. “Scaling Environmental Change through the Community-Level: A Trait-Based Response-and-Effect Framework for Plants.” Global Change Biology 14 (5): 1125–40. doi:10.1111/j.1365-2486.2008.01557.x. Templer, Pamela H. 2013. “Biogeochemistry: Limits on Carbon Uptake by Plants.” Nature Climate Change 3 (3): 184–85. doi:10.1038/nclimate1841. United Nations. 2012. “World Urbanization Prospects, The 2011 Revision”. United Nations publication. Weiss, Stuart B. 1999. “Cars, Cows, and Checkerspot Butterflies: Nitrogen Deposition 240 and Management of Nutrient-Poor Grasslands for a Threatened Species.” Conservation Biology 13 (6): 1476–86. doi:10.1046/j.1523-1739.1999.98468.x. Wentz, Elizabeth A., Patricia Gober, Robert C. Balling, and Thomas A. Day. 2002. “Spatial Patterns and Determinants of Winter Atmospheric Carbon Dioxide Concentrations in an Urban Environment.” Annals of the Association of American Geographers 92 (1): 15–28. doi:10.1111/1467-8306.00277. Zavaleta, Erika S., M. Rebecca Shaw, Nona R. Chiariello, Brian D. Thomas, Elsa E. Cleland, Christopher B. Field, and Harold A. Mooney. 2003. “GRASSLAND RESPONSES TO THREE YEARS OF ELEVATED TEMPERATURE, CO2, PRECIPITATION, AND N DEPOSITION.” Ecological Monographs 73 (4): 585–604. doi:10.1890/02-4053. 241 TABLE 1: CO2 and O3 concentrations in treatment chambers. Mean (SD) CO2 (ppm, 24 hour average) and O3 (ppb (0900 – 1800 daytime average) concentrations in treatment chambers (averaged within a chamber for 24 hour period) and within treatments (n = 3 chambers averaged per treatment) during the study period. Plus signs indicate ambient (+), mid (++), and high (+++) fumigation treatments. Carbon'dioxide''(ppm) Ozone'(ppb)' Individual'chamber + 419.5'(26.3) + 15.2'(7.4) + 409.7'(18.9) ++ 60.1'(22.0) + 410.4'(18.9) +++ 93.4'(21.2) ++ 530.3'(42.9) + 16.9'(9.4) ++ 537.2'(45.2) ++ 54.5'(14.4) ++ 550.2'(45.9) +++ 100.3'(30.1) +++ 666.1'(53.4) + 13.4'(8.9) +++ 650.6'(96.4) ++ 48.3'(23.0) +++ 660.9'(60.9) +++ 91.8'(25.9) Treatment + 413.2'(22.1) + 15.2'(8.7) ++ 539.4'(45.4) ++ 54.3'(20.9) +++ 659.1'(73.2) +++ 95.3'(26.3) 242 TABLE 2: Linear mixed model results showing individual, additive or interactive effects on growth rate and photosynthetic yield. Linear mixed model results showing elevated CO2, O3, and N as individual, additive or interactive effects on growth rate (mg day-1) and photosynthetic yield for non-native S. arabicus and native P. recurvata species. Akaike’s information critereon value for finite samples (AICc) and delta (AICc) are reported as goodness of fit for each model. Lowest AICc and delta AICc (< 3) are italicized to indicate the best explanatory models, and the selected best fit model (chosen based on likelihood ratio comparison test) is bolded. Non7native&S..arabicus Native&P..recurvata Growth&Rate Yield Growth&Rate Yield AICc Δ&AICc AICc Δ&AICc AICc Δ&AICc AICc Δ&AICc Null$model$(intercept) 163.6 14.5 1141.5 12.8 258.7 9.1 1130.7 5.3 Individual&effects Carbon$dioxide !76.2 1.9 1144.2 10.1 262.1 12.5 1127.7 8.3 Ozone 160.1 18 1141.4 12.9 249.6 0 !135.1 0.9 Nitrogen 160.9 17.2 1141.9 12.4 262.9 13.3 1128.2 7.8 Multiple&additive&effects CO2$+$O3 !78.1 0 1145.5 8.8 251.5 1.9 '136 0 O3$+$N 157.3 20.8 1142.1 12.2 254 4.4 1132.2 3.8 CO2$+$N 173.3 4.8 1146.3 8 266.4 16.8 1124.8 11.2 CO2$+$O3$+$N 175.1 3 1148.4 5.9 256 6.4 1132.5 3.5 Interaction&effects CO2$*$O3 '78.1 0 '154.3 0 258.1 8.5 !134.4 1.6 O3$*$N 151.8 26.3 1134.2 20.1 260.8 11.2 1122.4 13.6 CO2$*$N 168.8 9.3 1138.8 15.5 273.8 24.2 1115.9 20.1 CO2$*$O3$*$N 154.4 23.7 293.4 43.8 158.9 77.1 CO2,$O3$and$N$(and$their$interactions)$are$included$as$fixed$effects$and$growth$chamber$is$included$as$ a$random$effect$in$all$models.$$ 243 FIGURE 1: CSTR chambers in UC Riverside greenhouse. Nine Continuously Stirred Tank Reactor (CSTR) chambers at UC Riverside used for the multi-factor dose response fumigation experiment to test the co-occurring effects of CO2, O3 and N. 244 12" *" * * *" )' 81 9" Ozone'(ppb)' Ambient"(15)"" *" Mid"(60)"" 6" High"(100)"" 3" Growth'Rate'(mg'day **" 0" Updated" Ambient" Mid" High" data"with" 0.4" (400)"" (550)"" (700)"" Ozone+(ppb)+ Outliers" removed." Carbon'Dioxide'(ppm)' Ambient"(15)"" ********" Figure'XX:"Average"(+/@"1SE)"growth"rate"(mg"day@1)"for"S.#Mid"(60)""arabicus"in" USE"THIS" 0.3" GRAPH" combined"CO2"and"O3"treatments.""Asterisks"indicate"means"High"(100)"" significantly"different"from"control"based"on"*" DunneR"comparison"with" a"Bonferroni"adjustment.""Dashed"horizontal"line"represents"average" growth"rate"from"control"(grown"in"ambient"CO0.2" * 2"and"ambient"O3)"" *" Photosynthe)c+Yield+ 0.1" 0" Ambient" Mid" High" (400)"" (550)"" (700)"" Carbon+Dioxide+(ppm)+ Figure+2:"Average"(+/@"1SE)"growth"rate"(mg"day@1,"top)"and"photosyntheIc" yield"(boLom)"for"S.#arabicus"in"single"and"combined"CO2"and"O3" FIGURE 2:treatments."Dashed"horizontal"line"represents"control"average"(ambient" Growth (mg day-1) and photosynthetic yield of S. arabicus grown in ambient CO "and"ambient"O ).""Asterisks"indicate"significantly"different"means"from"-1 and elevated CO2 2 and O3. Average3 (+/- 1SE growth rate (mg day , top) and control"based"on"DunneL’s"comparison"with"Bonferroni"adjustment."" photosynthetic yield (unitless, bottom) for S. arabicus in single and combined CO2 and O3 treatments. Dashed horizontal line represents control average (ambient CO2 and ambient O3). Asterisks indicate significantly different means from control based on Dunnett’s comparison with Bonferroni adjustment. 245 8( )' 81 6( Ozone'(ppb)' Ambient((15)(( Mid((60)(( 4( High((100)(( *( 2( Growth'Rate'(mg'day 0( Ambient( Mid( High( **" *" Updated" (400)(( (550)(( (700)((*" Ozone'(ppb)' data"with" 0.4" Ambient"(15)"" Outliers" Carbon'Dioxide'(ppm)' removed." Mid"(60)"" Figure'XX:(Average((+/>(1SE)(growth(rate((mg(day>1)(for(Pectocarya) ********" High"(100)"" USE"THIS" recurvata(in(combined(CO2(and(O3(treatments.((Asterisk(indicates(mean( GRAPH" significantly(different(from(control(based(on(0.3" DunneQ(comparison(with(a( Bonferroni(adjustment.((Dashed(horizontal(line(represents(average( growth(rate(from(control((grown(in(ambient(CO2(and(ambient(O3).(( 0.2" Photosynthe6c'Yield' 0.1" 0" Ambient" Mid" High" (400)"" (550)"" (700)"" Carbon'Dioxide'(ppm)' Figure'3:"Average"(+/O"1SE)"growth"rate"(mg"dayO1,"top)"and"photosyntheRc" -1 FIGURE 3:yield"(boTom)"for" Growth (mg dayP.#recurvata) and photosynthetic"in"single"and"combined"CO yield of P. recurvata"and"O " grown in 2 3 -1 ambient andtreatments."Dashed"horizontal"line"represents"control"average"(ambient" elevated CO2 and O3. Average (+/- 1SE growth rate (mg day , top) and photosyntheticCO2"and"ambient"O yield (unitless,3).""Asterisks"indicate"significantly"different"means"from" bottom) for P. recurvata in single and combined CO2 and O3 treatments.control"based"on" Dashed horizontalDunneT’s line"comparison"with" represents controlBonferroni average"adjustment.""" (ambient CO2 and ambient O3). Asterisks indicate significantly different means from control based on Dunnett’s comparison with Bonferroni adjustment. 246 100#100$ Non'na,ve#graminoid#(S.#arabicus)# 75#75$ Na,ve#forb#(P.#recurvataSchismus$)# Pectocarya$ ' 50#50$ 25#25$ 0#0$ Percent'Change '25#!25$ '50#!50$ '75#!75$ CO2$ O3$ N$ CO2$x$O3$ CO2$x$N$ O3$x$N$ CO2$x$O3$ x$N$ FIGUREFigure'4 4: Observed:#Observed#percent#rela,ve#change#in#growth#rate#(compared# percent relative change in growth rate. Observed percent relative change in growth rate (compared to the control defined as ambient CO2, O3, and/or N for to#the#control#defined#as#ambient#CO2,#O3,#and/or#N#for#each# each treatmenttreatment#combina,on)#for#non'na,ve# combination) for non-native graminoidgraminoid (S. arabicus#(S.#arabicus, dark grey),#dark# and native forb (P. grey)#and#na,ve#forb#(recurvata, light grey) inP.# singrecurvatale and combined,#light#grey)#in#single#and#combined# high CO2 (700 ppm), O3 (100 ppb), and N (8 kg N ha-1 yr-1) treatments. high#CO (700ppm),#O (100ppb),#and#N#(8#kgN#ha'1#yr'1)#treatments.### 2# 3# 247 CO2$x$O3$ 75$ CO2$x$N$ O3$x$N$ Na)ve$forb$CO2$x$O3$x$N$ Non3na)ve$ (P.#recurvata) $ graminoid$ Observed( 25$ (S.#arabicus)$ !75$ !25$ 25$ 75$ Observed( !25$ CO2$x$O3$ CO2$x$N$ Predicted( O $x$N !75$ 3 $ CO $x$O $x$N Predicted( 2 3 $ Figure(5:$Predicted$and$observed$percent$change$in$growth$rate$(from$control$ FIGUREdefined$as$ambient$CO 5: Additive and ,$Onon,$and/or$N$for$each$treatment$combina)on).$$Predicted$-additive effects based on predicted and observed percent relative change in growth 2rate3 of S. arabicus and P. recurvata. Predicted and observed responses$are$the$sum$of$percent$change$from$individual$treatments$(CO2$+$O3,$CO2$ percent+$N,$ changeetc)$and$observed$responses$are$from$empirically$combined$treatments.$$ in growth rate (from the control defined as ambient CO2, O3, and/or N for eachCombina)ons$shown$for$high$(black)$and$middle$(dark$gray)$treatments$for$ treatment combination). Predicted responses are the sum of percent changeS.# from individualarabicus treatments#and$high$(light$gray)$and$middle$(open)$treatments$for$ (CO2 + O3, CO2 + N, etc) and observed responseP.#recurvatas are from.$Points$ empiricallyalong$the$dashed$1:1$line$indicate$addi)ve$responses,$below$the$line$are$antagonis)c$ combined treatments. Combinations shown for high (black) and middle (dark grey)effects,$and$above$the$line$are$synergis)c$effects.$$$$ treatments for S. arabicus and high (light grey) and middle (open) treatments for P. recurvata. Points along the dashed 1:1 line indicate additive responses, below the line are antagonistic effects and above the line are synergistic effects. Ellipses highlight the different species responses. 248 APPENDIX 1 TABLE 1: Multi-factor dose response treatment combinations within each chamber: O3 and CO2 were delivered by gas fumigation for desired concentrations within each Table 1: Multi-factor dose response treatment chamber and N was applied as three levels of NH4-NO3 within each chamber. combinations: O and CO fumigation and N applied as 3 2 NH -NO within each chamber. 4 3 Chamber O3 (ppb) N (kg/ha/yr) CO2 (ppm) 1: (Control) Ambient (~15) 0, 4, 8 Ambient (~400) 2: 15 0, 4, 8 550 3: 15 0, 4, 8 700 4: 60 0, 4, 8 400 5: 0, 4, 8 60 550 6: 60 0, 4, 8 700 7: 100 0, 4, 8 400 8: 100 0, 4, 8 550 9: 100 0, 4, 8 700 249 CHAPTER 6 CONCLUDING REMARKS In this conclusion, I synthesize the main research findings from each chapter and highlight the major contributions of this research, and then point to next steps and management implications. SYNTHESIS OF MAJOR RESEARCH FINDINGS Chapter 2: Examining the drivers of people’s actions and in turn the ecological outcome of those actions in highly managed systems, I found that a complex set of individual and institutional social characteristics drives people’s choices, which in turn affect ecological structure and functioning across scales from yards to cities (Larson et al. 2010; Cook, Hall, and Larson 2012). This work demonstrates the link between individuals’ decision- making and ecosystem service provisioning in highly managed urban ecosystems. In addition, this research highlights the importance of an interdisciplinary and multi-scalar approach to understanding complex human-ecological systems. Chapter 3: Overall, total nitrogen (N) deposition in the Phoenix metropolitan region and surrounding Sonoran Desert was lower (7.3 kg N ha-1 yr-1) than expected compared to similar ecosystems, but within the range of the dryland N critical load. The dryland critical load, estimated between 3 – 8 kg N ha-1 yr-1, is the level at which ecological changes are expected to occur as a result of elevated N inputs (Fenn et al. 2010; Pardo et 250 al. 2011). Despite low regional estimated deposition rates, N inputs from arid cities into the surrounding native ecosystem are likely to have significant ecological consequences, particularly with long-term accumulation and in years of above average precipitation. Using multiple sampling approaches to address the uncertainties related to quantifying N deposition in dryland systems, I found that dry N deposition was highest within the urban region, and primarily predicted by proximity to sources of N emissions. In contrast, patterns of total and wet deposition were mainly driven by meteorological variables and the timing and frequency of rainfall events. My results highlight the importance of employing multiple measurement techniques to accurately assess seasonal trends of wet and dry N deposition and their distribution across the landscape. Chapter 4: This study is the first to identify the distinct spatial pattern of co-occurring, ecologically important pollutants within protected lands. The urban “ecological airshed,” is created by elevated concentrations and deposition of biologically important urban- generated compounds. I found that the “ecological airshed” extends far into the outlying native ecosystem at levels that are likely to affect ecosystem structure and function. The impacts of the urban atmosphere on ecosystems, in combination with other factors related to urbanization such as elevated temperatures, may lead to longer-term consequences for ecosystem structure, functioning, and services. My findings show the need for air quality monitoring with an expanded repertoire of compounds known to affect ecosystem services and biological processes in both urban and remote native ecosystems. 251 Chapter 5: Under current and predicted future air quality conditions, herbaceous desert species responded differently suggesting potential cascading long-term consequences for community composition in native ecosystems. The non-native species (Schismus arabicus) grew rapidly in all treatments despite physiological stress under high O3 concentrations. In contrast, the native species (Pectocarya recurvata) was more sensitive to O3 and, unlike the non-native species, did not experience the protective cover of elevated CO2. Elevated CO2 mitigated negative effects of O3 for the non-native species and suggests that S. arabicus may be favored over P. recurvata when exposed to future air quality conditions. In addition, species responses were synergistic in combinations of CO2, O3, and N. Overall, these results provide empirical evidence for future nuanced policies addressing ecosystem sensitivity to multiple pollutants in native landscapes. NEXT STEPS AND IMPLICATIONS FOR FUTURE MANAGEMENT Global urban land area is expected to triple by 2030 and more than 65% of the world’s population is predicted to live in cities by 2050 (United Nations 2012; Seto, Güneralp, and Hutyra 2012). Human activities concentrated in growing urban centers are significant sources of atmospheric pollutants – including CO2, reactive N, and O3 – that affect air quality, human health, ecosystem services, and ecological functioning (IPCC 2014). My findings highlight that the urban “ecological airshed” extends well beyond the urban boundary where ecologically relevant pollutants co-occur in remote areas at levels that likely affect ecosystem structure and function. I also found that existing urban air quality conditions can affect growth and physiological functioning of primary producers 252 with potential long-term feedbacks on native ecosystem structure, function, and the provision of ecosystem services (Gregg, Jones, and Dawson 2003; Gregg, Jones, and Dawson 2006). Identifying the extent of land area exposed to human-generated pollutants in an urban “ecological airshed” highlights the need and urgency to adopt air quality monitoring and regulations that include a full suite of ecologically relevant compounds to protect the surrounding natural environments and ecosystem services. Neither current air quality nor conservation policies typically account for the realistic scenarios in which multiple urban pollutants co-occur with potential synergistic or antagonistic impacts on outlying ecosystems and human health (Clair et al. 2011). Air- quality regulations with a primary focus on human health and secondarily on visibility and ecosystem impacts. Thus, air quality regulations mainly focus on multiple pollutants only with regard to atmospheric chemical reactions between precursors and pollutants of interest for human health (e.g. VOC, NOx, and O3; Finlayson-Pitts & Pitts 2000). While conservation plans seek to preserve native ecosystems, they frequently do not account for the influence of air pollutants that cross political and municipal boundaries (Lovett et al. 2009). In order to more effectively manage and address air quality concerns for humans and ecosystems, there are many changes to current regulations that are needed. For example, new regulatory policies should 1) consider a more inclusive suite of pollutants that affect humans and ecosystems, 2) monitor ambient concentrations in areas of human and ecosystem exposure, 3) model projected changes in air quality under different scenarios, 4) regulate emissions, and 5) test and account for ecological and human exposure, risks, and responses to multiple pollutants (Hidy and Pennell 2010). 253 Critical loads are an example framework developed in the European Union and United States that account for ecosystem exposure and risk. To date, critical loads have been determined only for single pollutants, such as N deposition, rather than accounting for multiple, often more realistic exposure of ecosystems to co-occurring compounds. Similarly, air quality regulations focused on human health address individual pollutants, such as O3 or particulate matter. I propose that a multi-pollutant critical load framework would be an effective tool to regulate and examine both ecosystem and human exposure to multiple relevant stressors. This would be an important step in addressing the resulting environmental and human health impacts from air quality conditions. My research highlights the extent of the “ecological airshed” and that existing urban air quality conditions can alter ecosystem structure and functioning in nearby native ecosystems. These findings on multiple interacting pollutants provide initial empirical evidence and first steps for establishing nuanced regulations to protect ecosystems and human health through a multi-pollutant critical load. 254 REFERENCES Clair, Thomas A., Douglas Burns, Irma Rosas Pérez, Jules Blais, and Kevin Percy. 2011. “Ecosystems.” In Technical Challenges of Multipollutant Air Quality Management, edited by George M. Hidy, Jeffrey R. Brook, Kenneth L. Demerjian, Luisa T. Molina, William T. Pennell, and Richard D. Scheffe, 139– 229. Springer Netherlands. http://link.springer.com/chapter/10.1007/978-94-007- 0304-9_6. Cook, Elizabeth M., Sharon J. Hall, and Kelli L. Larson. 2012. “Residential Landscapes as Social-Ecological Systems: A Synthesis of Multi-Scalar Interactions between People and Their Home Environment.” Urban Ecosystems 15 (1): 19–52. doi:10.1007/s11252-011-0197-0. Fenn, Mark E., E. B. Allen, S. B. Weiss, S. Jovan, L. H. Geiser, G. S. Tonnesen, R. F. Johnson, et al. 2010. “Nitrogen Critical Loads and Management Alternatives for N-Impacted Ecosystems in California.” Journal of Environmental Management 91 (12): 2404–23. doi:10.1016/j.jenvman.2010.07.034. Gregg, Jillian W., Clive G. Jones, and Todd E. Dawson. 2003. “Urbanization Effects on Tree Growth in the Vicinity of New York City.” Nature 424 (6945): 183–87. doi:10.1038/nature01728. ———. 2006. “PHYSIOLOGICAL AND DEVELOPMENTAL EFFECTS OF O3 ON COTTONWOOD GROWTH IN URBAN AND RURAL SITES.” Ecological Applications 16 (6): 2368–81. doi:10.1890/1051- 0761(2006)016[2368:PADEOO]2.0.CO;2. IPCC. 2014. “Intergovernmental Panel on Climate Change: Fifth Assessment Report.” http://www.ipcc.ch/. Larson, Kelli L., Elizabeth Cook, Colleen Strawhacker, and Sharon J. Hall. 2010. “The Influence of Diverse Values, Ecological Structure, and Geographic Context on Residents’ Multifaceted Landscaping Decisions.” Human Ecology 38 (6): 747–61. doi:10.1007/s10745-010-9359-6. Lovett, Gary M., Timothy H. Tear, David C. Evers, Stuart E.G. Findlay, B. Jack Cosby, Judy K. Dunscomb, Charles T. Driscoll, and Kathleen C. Weathers. 2009. “Effects of Air Pollution on Ecosystems and Biological Diversity in the Eastern United States.” Annals of the New York Academy of Sciences 1162 (1): 99–135. doi:10.1111/j.1749-6632.2009.04153.x. Pardo, Linda H., Mark E. Fenn, Christine L. Goodale, Linda H. Geiser, Charles T. 255 Driscoll, Edith B. Allen, Jill S. Baron, et al. 2011. “Effects of Nitrogen Deposition and Empirical Nitrogen Critical Loads for Ecoregions of the United States.” Ecological Applications 21 (8): 3049–82. doi:10.1890/10-2341.1. Seto, Karen C., Burak Güneralp, and Lucy R. Hutyra. 2012. “Global Forecasts of Urban Expansion to 2030 and Direct Impacts on Biodiversity and Carbon Pools.” Proceedings of the National Academy of Sciences 109 (40): 16083–88. doi:10.1073/pnas.1211658109. United Nations. 2012. “World Urbanization Prospects, The 2011 Revision”. United Nations publication. 256 REFERENCES CHAPTER 1 Bettencourt, Luís M. A. 2013. “The Origins of Scaling in Cities.” Science 340 (6139): 1438–41. doi:10.1126/science.1235823. Bobbink, Roland, K. Hicks, J. Galloway, T. Spranger, R. Alkemade, M. Ashmore, M. Bustamante, S. Cinderby, E. Davidson, and F. Dentener. 2010. “Global Assessment of Nitrogen Deposition Effects on Terrestrial Plant Diversity: A Synthesis.” Ecological Applications 20 (1): 30–59. Boone, Christopher G., Elizabeth Cook, Sharon J. Hall, Marcia L. Nation, Nancy B. Grimm, Carol B. Raish, Deborah M. Finch, and Abigail M. York. 2012. “A Comparative Gradient Approach as a Tool for Understanding and Managing.” Urban Ecosystems 15 (4): 795–807. doi:10.1007/s11252-012-0240-9. Galloway, James N., Alan R. Townsend, Jan Willem Erisman, Mateete Bekunda, Zucong Cai, John R. Freney, Luiz A. Martinelli, Sybil P. Seitzinger, and Mark A. Sutton. 2008. “Transformation of the Nitrogen Cycle: Recent Trends, Questions, and Potential Solutions.” Science 320 (5878): 889–92. Grimm, Nancy B., Stanley H. Faeth, Nancy E. Golubiewski, Charles L. Redman, Jianguo Wu, Xuemei Bai, and John M. Briggs. 2008. “Global Change and the Ecology of Cities.” Science 319 (5864): 756–60. doi:10.1126/science.1150195. Grimm, Nancy B., David Foster, Peter Groffman, J. Morgan Grove, Charles S. Hopkinson, Knute J. Nadelhoffer, Diane E. Pataki, and Debra PC Peters. 2008. “The Changing Landscape: Ecosystem Responses to Urbanization and Pollution across Climatic and Societal Gradients.” Frontiers in Ecology and the Environment 6 (5): 264–72. Grimm, Nancy B., J. M. Grove, S. T. A. Pickett, and C. L. Redman. 2000. “Integrated Approaches to Long-Term Studies of Urban Ecological Systems.” Bioscience 50 (7): 571–84. doi:10.1641/0006-3568(2000)050[0571:IATLTO]2.0.CO;2. Hall, Sharon J., Ryan A. Sponseller, Nancy B. Grimm, David Huber, Jason P. Kaye, Christopher Clark, and Scott L. Collins. 2011. “Ecosystem Response to Nutrient Enrichment across an Urban Airshed in the Sonoran Desert.” Ecological Applications 21 (3): 640–60. doi:10.1890/10-0758.1. Hidy, George M., and William T. Pennell. 2010. “Multipollutant Air Quality Management.” Journal of the Air & Waste Management Association 60 (6): 645– 257 74. doi:10.3155/1047-3289.60.6.645. IPCC. 2014. “Intergovernmental Panel on Climate Change: Fifth Assessment Report.” http://www.ipcc.ch/. MEA. 2005. “Millennium Ecosystem Assessment: Ecosystems and Human Well-Being.” Island Press. http://www.unep.org/maweb. Pickett, S. T. A., M. L. Cadenasso, J. M. Grove, C. H. Nilon, R. V. Pouyat, W. C. Zipperer, and R. Costanza. 2001. “Urban Ecological Systems: Linking Terrestrial Ecological, Physical, and Socioeconomic Components of Metropolitan Areas.” Annual Review of Ecology and Systematics 32: 127–57. doi:10.1146/annurev.ecolsys.32.081501.114012. Seto, Karen C., Burak Güneralp, and Lucy R. Hutyra. 2012. “Global Forecasts of Urban Expansion to 2030 and Direct Impacts on Biodiversity and Carbon Pools.” Proceedings of the National Academy of Sciences 109 (40): 16083–88. doi:10.1073/pnas.1211658109. Smith, Melinda D., Alan K. Knapp, and Scott L. Collins. 2009. “A Framework for Assessing Ecosystem Dynamics in Response to Chronic Resource Alterations Induced by Global Change.” Ecology 90 (12): 3279–89. doi:10.1890/08-1815.1. United Nations. 2012. “World Urbanization Prospects, The 2011 Revision”. United Nations publication. Vitousek, Peter M., John D. Aber, Robert W. Howarth, Gene E. Likens, Pamela A. Matson, David W. Schindler, William H. Schlesinger, and David G. Tilman. 1997. “HUMAN ALTERATION OF THE GLOBAL NITROGEN CYCLE: SOURCES AND CONSEQUENCES.” Ecological Applications 7 (3): 737–50. doi:10.1890/1051-0761(1997)007[0737:HAOTGN]2.0.CO;2. Zavaleta, Erika S., M. Rebecca Shaw, Nona R. Chiariello, Brian D. Thomas, Elsa E. Cleland, Christopher B. Field, and Harold A. Mooney. 2003. “GRASSLAND RESPONSES TO THREE YEARS OF ELEVATED TEMPERATURE, CO2, PRECIPITATION, AND N DEPOSITION.” Ecological Monographs 73 (4): 585–604. doi:10.1890/02-4053. CHAPTER 2 Abraham A, Sommerhalder K, Abel T (2010) Landscape and well-being: A scoping study on the health-promoting impact of outdoor environments. Int J Public Health 55 (1):59-69 258 Acar C, Acar H, Eroglu E (2007) Evaluation of ornamental plant resources to urban biodiversity and cultural changing: A case study of residential landscapes in Trabzon City (Turkey). Build Environ 42 (1):218-229 Ajzen I (1985) From intentions to actions: A theory of planned behavior. In: Kuhl J, Beckmann J (eds) Action control, from cognition to behavior. Springer series in social psychology. Springer-Verlag, New York, pp 11-39 Akinnifesi F, Sileshi G, Ajayi O, Akinnifesi A, de Moura E, Linhares J, Rodrigues I (2010) Biodiversity of the urban homegardens of São Luís City, northeastern Brazil. Urban Ecosyst 13 (1):129-146 Albuquerque UP, Andrade LHC, Caballero J (2005) Structure and floristics of homegardens in northeastern Brazil. J Arid Environ 62 (3):491-506 Alp S, Ozturk S, Turkoglu N, Koyuncu M (2010) Basic elements of the traditional garden identity in the city of Van. Afr J Agric Res 5 (11):1277-1283 Alumai A, Grunkemeyer M, Kovach J, Shetlar D, Cardina J, Rimelspach J, Clayton S, Grewal P (2010) Implementing integrated pest management in professional lawn care: A case study. Urban Ecosyst 13 (1):37-49 Alumai A, Salminen S, Richmond D, Cardina J, Grewal P (2009) Comparative evaluation of aesthetic, biological, and economic effectiveness of different lawn management programs. Urban Ecosyst 12 (2):127-144 Andersson E, Barthel S, Ahrne K (2007) Measuring social-ecological dynamics behind the generation of ecosystem services. Ecol Appl 17 (5):1267-1278 Askew LE, McGuirk PM (2004) Watering the suburbs: Distinction, conformity and the suburban garden. Aust Geographer 35 (1):17-37 Aurora A, Simpson T, Small M, Bender K (2009) Toward increasing avian diversity: Urban wildscapes programs. Urban Ecosyst 12 (3):347-358 Baker LA, Hartzheim PM, Hobbie SE, King JY, Nelson KC (2007) Effect of consumption choices on fluxes of carbon, nitrogen and phosphorus through households. Urban Ecosyst 10 (2):97-117 Baker PJ, Harris S (2007) Urban mammals:What does the future hold? An analysis of the factors affecting patterns of use of residential gardens in Great Britain. Mamm Rev 37 (4):297-315 259 Balling RC, Gober P (2007) Climate variability and residential water use in the city of Phoenix, Arizona. J Appl Meteorol Climatol 46 (7):1130-1137 Band LE, Cadenasso ML, Grimmond CS, Grove JM, Pickett STA (2005) Heterogeneity in urban ecosystems: Patterns and process. In: Lovett GM, Turner MG, Jones CG, Weathers KC (eds) Ecosystem function in heterogeneous landscapes. Springer, New York, pp 257-278 Baris ME, Sahin S, Yazgan ME (2009) The contribution of trees and green spaces to the urban climate: The case of Ankara. Afr J of Agric Res 4 (9):791-800 Beard JB, Green RL (1994) The role of turfgrass in environemntal protection and their benefits to humans. J Environ Qual 23 (3) Bennett EM, Carpenter SR, Clayton MK (2005) Soil phosphorus variability: Scale- dependence in an urbanizing agricultural landscape. Landsc Ecol 20 (4):389-400 Bhatti M, Church A (2001) Cultivating natures: Homes and gardens in late modernity. Sociol 35 (02):365-383 Bijoor NS, Czimczik CI, Pataki DE, Billings SA (2008) Effects of temperature and fertilization on nitrogen cycling and community composition of an urban lawn. Glob Change Biol 14 (9):2119-2131 Blanchoud H, Farrugia F, Mouchel JM (2004) Pesticide uses and transfers in urbanised catchments. Chemosph 55 (6):905-913 Bonan GB (2000) The microclimates of a suburban Colorado (USA) landscape and implications for planning and design. Landsc Urban Plan 49 (3-4):97-114 Boone CG, Cadenasso ML, Grove JM, Schwarz K, Buckley GL (2010) Landscape, vegetation characteristics, and group identity in an urban and suburban watershed: Why the 60s matter. Urban Ecosyst 13:255-271 Bormann FH, Balmori D, Geballe GT (2001) Redesigning the American lawn: A search for environmental harmony. 2nd edition. Yale University Press, New Haven, CT Bowman T, Thompson J (2009) Barriers to implementation of low-impact and conservation subdivision design: Developer perceptions and resident demand. Landsc Urban Plan 92 (2):96-105 Braman SK, Latimer JG, Oetting RD, McQueen RD, Eckberg TB, Prinster M (2000) Management strategy, shade, and landscape composition effects on urban landscape plant quality and arthropod abundance. J Econ Entomol 93 (5):1464- 1472 260 Braman SK, Oetting RD, Florkowski W (1997) Assessment of pesticide use by commercial landscape maintenance and lawn care firms in Georgia. J Entomol Sci 32 (4):403-411 Burghardt KT, Tallamy DW, Shriver WG (2009) Impact of native plants on bird and butterfly biodiversity in suburban landscapes. Conserv Biol 23 (1):219-224 Buyantuyev A, Wu JG (2010) Urban heat islands and landscape heterogeneity: Linking spatiotemporal variations in surface temperatures to land-cover and socioeconomic patterns. Landsc Ecol 25:17-33 Byrne LB (2006) Effects of of urban habitat types and landscape patterns on ecological variables at the aboveground–belowground interface. PhD Dissertation, Pennsylvania State University Byrne LB (2007) Habitat structure: A fundamental concept and framework for urban soil ecology. Urban Ecosyst 10 (3):255-274 Byrne LB, Bruns MA (2004) The effects of lawn management on soil microarthropods. J Agr Urban Entomol 21 (3):150-156 Byrne LB, Bruns MA, Kim KC (2008) Ecosystem properties of urban land covers at the aboveground-belowground interface. Ecosyst 11 (7):1065-1077 Byrne LB, Grewal P (2008) Introduction to ecological landscaping: A holistic description and framework to guide the study and management of urban landscape parcels. CATE 1 (2) Catterall CP, Cousin JA, Piper S, Johnson G (2010) Long-term dynamics of bird diversity in forest and suburb: Decay, turnover or homogenization? Divers Distrib 16 (4):559-570 Chace JF, Walsh JJ (2006) Urban effects on native avifauna: A review. Landsc Urban Plan 74 (1):46-69 Chamberlain DE, Cannon AR, Toms MP (2004) Associations of garden birds with gradients in garden habitat and local habitat. Ecography 27 (5):589-600 Cheng Z, Richmond D, Salminen S, Grewal P (2008) Ecology of urban lawns under three common management programs. Urban Ecosyst 11 (2):177-195 Cheshire LA, Rosenblatt T, Lawrence G, Walters P (2009) The governmentality of master planning: Housing consumption, aesthetics and community on a new estate. Hous Stud 24 (5):653-667 261 Clayton S (2007) Domesticated nature: Motivations for gardening and perceptions of environmental impact. J Environ Psychol 27 (3):215-224 Coiacetto E (2007) The role of the development industry in shaping urban social space: A conceptual model. Geog Res 45 (4):340-347 Collins JP, Kinzig A, Grimm NB, Fagan WF, Hope D, Wu JG, Borer ET (2000) A new urban ecology. Am Sci 88 (5):416-425 Conway T (2009) Local environmental impacts of alternative forms of residential development. Environ Plan B-Plan Des 36 (5):927-943 Cook WM, Faeth SH (2006) Irrigation and land use drive ground arthropod community patterns in an urban desert. Environ Entomol 35 (6):1532-1540 Corral-Verdugo V, Frias-Amenta M, Gonzalez-Lomeli D (2003) On the relationship between anti-social and anti-environmental behaviors: An empirical study. Popul Environ 24 (3):273-286 Corral-Verdugo V, Frias-Armenta M, Perez-Urias F, Orduna-Cabrera V, Espinoza- Gallego N (2002) Residential water consumption, motivation for conserving water and the continuing tragedy of the commons. Environ Manag 30 (4):527-535 Coutts AM, Beringer J, Tapper NJ (2007) Impact of increasing urban density on local climate: Spatial and temporal variations in the surface energy balance in Melbourne, Australia. J of Appl Meteorol Climatol 46 (4):477-493 Crawford SES, Ostrom E (1995) A grammar of institutions. Am Political Sci Rev 89 (3):582-600 Crow T, Brown T, De Young R (2006) The Riverside and Berwyn experience: Contrasts in landscape structure, perceptions of the urban landscape, and their effects on people. Landsc Urban Plan 75 (3-4):282-299 Curtis KR, Cowee MW (2010) Are homeowners willing to pay for “origin-certified" plants in water-conserving residential landscaping? J Agric Resour Econ 35 (1):118-132 Daniels GD, Kirkpatrick JB (2006a) Comparing the characteristics of front and back domestic gardens in Hobart, Tasmania, Australia. Landsc Urban Plan 78 (4):344- 352 Daniels GD, Kirkpatrick JB (2006b) Does variation in garden characteristics influence the conservation of birds in suburbia? Biol Conserv 133 (3):326-335 262 Davies R, Hall SJ (2010) Direct and indirect effects of urbanization on soil and plant nutrients in desert ecosystems of the Phoenix metropolitan area, Arizona (USA). Urban Ecosyst 13 (3):295-317 Davies ZG, Fuller RA, Loram A, Irvine KN, Sims V, Gaston KJ (2009) A national scale inventory of resource provision for biodiversity within domestic gardens. Biol Conserv 142 (4):761-771 Dietz T, Fitzgerald A, Shwom R (2005) Environmental values. Annu Rev Environ Resour 30:335-372 Doody BJ, Sullivan JJ, Meurk CD, Stewart GH, Perkins HC (2010) Urban realities: The contribution of residential gardens to the conservation of urban forest remnants. Biodivers Conserv 19 (5):1385-1400 Dorney JR, Guntenspergen GR, Keough JR, Stearns F (1984) Composition and structure of an urban woody plant community. Urban Ecol 8 (1-2):69-90 Dow K (2000) Social dimensions of gradients in urban ecosystems. Urban Ecosyst 4 (4):255-275 Dowling R, Atkinson R, McGuirk P (2010) Privatism, privatisation and social distinction in master-planned residential estates. Urban Policy Res 28 (4):391-410 Dunlap RE, Jones RE (2002) Environmental conern: Conceptual and measurement issues. In: Dunlap RE, Michelson W (eds) Handbook of Environmental Sociology. Greenwood Press, Westport, CT, pp 482-542 Dyckman CS (2008) The covenant conundrum in urban water conservation. Urban Lawyer 40 (1):17-50 Edgar DR, Kershaw GP (1994) The density and diversity of the bird populations in three residential communities in Edmonton, Alberta. Can Field-Nat 108 (2):156-161 Effland WR, Pouyat RV (1997) The genesis, classification, and mapping of soils in urban areas. Urban Ecosyst 1 (4):217-228 Eichemberg MT, Amorozo MCD, de Moura LC (2009) Species composition and plant use in old urban homegardens in Rio Claro, southeast of Brazil. Acta Bot Bras 23 (4):1057-1075 Eigenbrode SD, O'Rourke M, Wulfhorst JD, Althoff DM, Goldberg CS, Merrill K, Morse W, Nielsen-Pincus M, Stephens J, Winowiecki L, Bosque-Perez NA (2007) Employing philosophical dialogue in collaborative science. Biosci 57 (1):55-64 263 Endter-Wada J, Kurtzman J, Keenan SP, Kjelgren RK, Neale CMU (2008) Situational waste in landscape watering: Residential and business water use in an urban Utah community. J Am Water Resour Assoc 44 (4):902-920 Erickson JE, Cisar JL, Volin JC, Snyder GH (2001) Comparing nitrogen runoff and leaching between newly established St. Augustine grass turf and an alternative residential landscape. Crop Sci 41 (6):1889-1895 Evans KL, Newson SE, Gaston KJ (2009) Habitat influences on urban avian assemblages. Ibis 151 (1):19-39 Faeth SH, Warren PS, Shochat E, Marussich WA (2005) Trophic dynamics in urban communities. Biosci 55 (5):399-407 Feagan RB, Ripmeester M (1999) Contesting natural(ized) lawns: A geography of private green space in the Niagara region. Urban Geogr 20 (7):617-634 Fetridge ED, Ascher JS, Langellotto GA (2008) The bee fauna of residential gardens in a suburb of New York City (Hymenoptera: Apoidea). Ann Entomol Soc Am 101 (6):1067-1077 Foster D, Swanson F, Aber J, Burke I, Brokaw N, Tilman D, Knapp A (2003) The importance of land-use legacies to ecology and conservation. Biosci 53 (1):77-88 French K, Major R, Hely K (2005) Use of native and exotic garden plants by suburban nectarivorous birds. Biol Conserv 121 (4):545-559 Fuller RA, Irvine KN, Devine-Wright P, Warren PH, Gaston KJ (2007) Psychological benefits of greenspace increase with biodiversity. Biol Lett 3 (4):390-394 Fuller RA, Warren PH, Armsworth PR, Barbosa O, Gaston KJ (2008) Garden bird feeding predicts the structure of urban avian assemblages. Divers Distrib 14 (1):131-137 Gaston KJ, Warren PH, Thompson K, Smith RM (2005) Urban domestic gardens (IV): The extent of the resource and its associated features. Biodivers Conserv 14 (14):3327-3349 Gavier-Pizarro GI, Radeloff VC, Stewart SI, Huebner CD, Keuler NS (2010) Housing is positively associated with invasive exotic plant species richness in new england, USA. Ecol Appl 20 (7):1913-1925 Georgi JN, Dimitriou D (2010) The contribution of urban green spaces to the improvement of environment in cities: Case study of Chania, Greece. Build Environ 45 (6):1401-1414 264 Germaine SS, Rosenstock SS, Schweinsburg RE, Richardson WS (1998) Relationships among breeding birds, habitat, and residential development in greater Tucson, Arizona. Ecol Appl 8 (3):680-691 Germaine SS, Schweinsburg RE, Germaine HL (2001) Effects of residential density on Sonoran desert nocturnal rodents. Urban Ecosyst 5 (3):179-185 Gober P, Trapido-Lurie B (2006) Metropolitan Phoenix: Place making and community building in the desert. Metropolitan Portraits. University of Pennsylvania Press, Philadelphia, PA Gobster PH, Nassauer JI, Daniel TC, Fry G (2007) The shared landscape: What does aesthetics have to do with ecology? Landsc Ecol 22 (7):959-972 Goddard MA, Dougill AJ, Benton TG (2010) Scaling up from gardens: Biodiversity conservation in urban environments. Trends Ecol Evol 25 (2):90-98 Goffman E (1959) The presentation of self in everyday life. Doubleday Anchor Books. Doubleday, Garden City, NY Golubiewski NE (2006) Urbanization increases grassland carbon pools: Effects of landscaping in Colorado's front range. Ecol Appl 16 (2):555-571 Green DM, Oleksyszyn M (2002) Enzyme activities and carbon dioxide flux in a Sonoran desert urban ecosystem. Soil Sci Soc of Am J 66:2002-2008 Grimm NB, Faeth SH, Golubiewski NE, Redman CL, Wu JG, Bai XM, Briggs JM (2008) Global change and the ecology of cities. Sci 319 (5864):756-760 Grimm NB, Grove JM, Pickett STA, Redman CL (2000) Integrated approaches to long- term studies of urban ecological systems. Biosci 50 (7):571-584 Groffman PM, Law NL, Belt KT, Band LE, Fisher GT (2004) Nitrogen fluxes and retention in urban watershed ecosystems. Ecosyst 7 (4):393-403 Groffman PM, Pouyat RV (2009) Methane uptake in urban forests and lawns. Environ Sci Technol 43 (14):5229-5235 Groffman PM, Williams CO, Pouyat RV, Band LE, Yesilonis ID (2009) Nitrate leaching and nitrous oxide flux in urban forests and grasslands. J Environ Qual 38 (5):1848-1860 265 Grove JM, Cadenasso ML, Burch WR, Pickett STA, Schwarz K, O'Neil-Dunne J, Wilson M, Troy A, Boone C (2006a) Data and methods comparing social structure and vegetation structure of urban neighborhoods in Baltimore, Maryland. Soc Nat Resour 19 (2):117-136 Grove JM, Troy AR, O'Neil-Dunne JPM, Burch WR, Cadenasso ML, Pickett STA (2006b) Characterization of households and its implications for the vegetation of urban ecosystems. Ecosyst 9 (4):578-597 Guhathakurta S, Gober P (2007) The impact of the Phoenix urban heat island on residential water use. J Am Plan Assoc 73 (3):317-329 Hall SJ, Ahmed B, Ortiz P, Davies R, Sponseller RA, Grimm NB (2009) Urbanization alters soil microbial functioning in the Sonoran desert. Ecosyst 12 (4):654-671 Hall SJ, Huber D, Grimm NB (2008) Soil N2O and NO emissions from an arid, urban ecosystem. J Geophys Res-Biogeosci 113 (G1) Hall T (2010) Goodbye to the backyard?: The minimisation of private open space in the Australian outer-suburban estate. Urban Policy Res 28 (4):411-433 Harlan SL, Brazel AJ, Prashad L, Stefanov WL, Larsen L (2006) Neighborhood microclimates and vulnerability to heat stress. Soc Sci Med 63 (11):2847-2863 Harlan SL, Yabiku ST, Larsen L, Brazel AJ (2009) Household water consumption in an arid city: Affluence, affordance, and attitudes. Soc Nat Resour 22 (8):691-709 He JJ, Jia B (2007) Sustainable residential landscapes: A case study in Guangzhou, China. Lands Res 32 (2):241-254 Henderson SPB, Perkins NH, Nelischer M (1998) Residential lawn alternatives: A study of their distribution, form and structure. Landsc Urban Plan 42 (2-4):135-145 Hill TD, Polsky C (2007) Suburbanization and drought: A mixed methods vulnerability assessment in rainy Massachusetts. Environ Hazards 7 (4):291-301 Hirsch R, Baxter J (2009) The look of the lawn: Pesticide policy preference and health- risk perception in context. Environ Plan C-Gov Policy 27 (3):468-490 Holland JH (1995) Hidden order: How adaptation builds complexity. Helix Books. Addison-Wesley, Reading, MA Hope D, Gries C, Casagrande D, Redman CL, Grimm NB, Martin C (2006) Drivers of spatial variation in plant diversity across the Central Arizona-Phoenix ecosystem. Soc Nat Resour 19 (2):101-116 266 Hope D, Gries C, Zhu WX, L. FW, Redman CL, Grimm NB, Nelson AL, Martin C, Kinzig AP (2003) Socioeconomics drive urban plant diversity. PNAS 100 (15):8788-8792 Hope D, Zhu W, Gries C, Oleson J, Kaye J, Grimm NB, Baker L (2005) Spatial variation in soil inorganic nitrogen across an arid urban ecosystem. Urban Ecosyst 8 (3):251-273 Hourigan CL, Catterall CP, Jones D, Rhodes M (2010) The diversity of insectivorous bat assemblages among habitats within a subtropical urban landscape. Austral Ecol 35 (8):849-857 Huai H, Hamilton A (2009) Characteristics and functions of traditional homegardens: A review. Front Biol Chin 4 (2):151-157 Huang YJ, Akbari H, Taha H, Rosenfeld AH (1987) The potential of vegetation in reducing summer cooling loads in residential buildings. J Clim Appl Meteorol 26 (9):1103-1116 Hurd BH (2006) Water conservation and residential landscapes: Household preferences, household choices. J Agric Resour Econ 31 (2):173-192 Hurd BH, St Hilaire R, White JM (2006) Residential landscapes, homeowner attitudes, and water-wise choices in New Mexico. HortTech 16 (2):241-246 Iverson LR, Cook EA (2000) Urban forest cover of the Chicago region and its relation to household density and income. Urban Ecosyst 4:104-124 Jackson KT (1985) Crabgrass frontier: The suburbanization of the United States. Oxford University Press, NY James P, Tzoulas K, Adams MD, et al. (2009) Towards an integrated understanding of green space in the European built environment. Urban For Urban Green 8 (2):65- 75 Jenerette GD, Harlan SL, Brazel A, Jones N, Larsen L, Stefanov WL (2007) Regional relationships between surface temperature, vegetation, and human settlement in a rapidly urbanizing ecosystem. Landsc Ecol 22 (3):353-365 Jenerette GD, Wu JG, Grimm NB, Hope D (2006) Points, patches, and regions: Scaling soil biogeochemical patterns in an urbanized arid ecosystem. Glob Change Biol 12 (8):1532-1544 267 Jenkins VS (1994) The lawn: A history of an American obsession. Smithsonian Institution Press, Washington, DC Jenny H (1941) Factors of soil formation: A system of quantitative pedology. McGraw- Hill, New York, London Jorgensen A, Hitchmough J, Dunnett N (2007) Woodland as a setting for housing appreciation and fear and the contribution to residential satisfaction and place identity in Warrington New Town, UK. Landsc Urban Plan 79:273–28 Kabir ME, Webb EL (2009) Household and homegarden characteristics in southwestern Bangladesh. Agrofor Syst 75 (2):129-145 Kaiser FG, Wolfing S, Fuhrer U (1999) Environmental attitude and ecological behaviour. J Environ Psychol 19 (1):1-19 Karr CJ, Solomon GM, Brock-Utne AC (2007) Health effects of common home, lawn, and garden pesticides. Pediatr Clin N Am 54 (1):63-80 Kaye JP, Burke IC, Mosier AR, Guerschman JP (2004) Methane and nitrous oxide fluxes from urban soils to the atmosphere. Ecol Appl 14 (4):975-981 Kaye JP, Majumdar A, Gries C, Buyantuyev A, Grimm NB, Hope D, Jenerette GD, Zhu WX, Baker L (2008) Hierarchical bayesian scaling of soil properties across urban, agricultural, and desert ecosystems. Ecol Appl 18 (1):132-145 Kaye JP, McCulley RL, Burke IC (2005) Carbon fluxes, nitrogen cycling, and soil microbial communities in adjacent urban, native and agricultural ecosystems. Glob Change Biol 11 (4):575-587 Kiely T, Donaldson D, Grube A (2004) Pesticide industry sales and usage: 2000 and 2001 market estimates. Environmental Protection Agency; Office of Pesticide Programs, Washington DC Kinzig AP, Warren P, Martin C, Hope D, Katti M (2005) The effects of human socioeconomic status and cultural characteristics on urban patterns of biodiversity. Ecol Soc 10 (1):23 Kirkpatrick J, Daniels G, Davison A (2009) An antipodean test of spatial contagion in front garden character. Landsc Urban Plan 93 (2):103-110 Kirkpatrick J, Daniels G, Zagorski T (2007) Explaining variation in front gardens between suburbs of Hobart, Tasmania, Australia. Landsc Urban Plan 79 (3- 4):314-322 268 Kjelgren R, Rupp L, Kilgren D (2000) Water conservation in urban landscapes. Hortsci 35 (6):1037-1040 Klingeman WE, Pettis GV, Braman SK (2009) Lawn care and landscape maintenance: Professional acceptance of insect- and disease-resistant ornamental plants. Hortsci 44 (6):1608-1615 Koerner B, Klopatek J (2002) Anthropogenic and natural CO2 emission sources in an arid urban environment. Environ Pollut 116:S45-S51 Kweon BS, Ellis CD, Leiva PI, Rogers GO (2010) Landscape components, land use, and neighborhood satisfaction. Environ Plan B-Plan Des 37 (3):500-517 Larsen L, Harlan SL (2006) Desert dreamscapes: Residential landscape preference and behavior. Landsc Urban Plan 78 (1-2):85-100 Larson KL, Casagrande D, Harlan SL, Yabiku ST (2009a) Residents' yard choices and rationales in a desert city: Social priorities, ecological impacts, and decision tradeoffs. Environ Manag 44 (5):921-937 Larson KL, Cook EM, Strawhacker CS, Hall SJ (2010) The influence of diverse values, ecological structure and geographic context on residents’ multifaceted landscaping decisions. Hum Ecol 38 (6):747-761 Larson KL, Gustafson A, Hirt P (2009b) Insatiable thirst and a finite supply: An assessment of municipal water-conservation policy in greater Phoenix, Arizona, 1980-2007. J Policy Hist 21 (02):107-137 Law NL, Band LE, Grove JM (2004) Nitrogen input from residential lawn care practices in suburban watersheds in Baltimore County, MD. J Environ Plan Manag 47 (5):737-755 Lehmann A, Stahr K (2007) Nature and significance of anthropogenic urban soils. J Soils Sediments 7 (4):247-260 Lewis DB, Kaye JP, Gries C, Kinzig AP, Redman CL (2006) Agrarian legacy in soil nutrient pools of urbanizing arid lands. Glob Change Biol 12:703-709 Liu HB, Hull RJ, Duff DT (1997) Comparing cultivars of three cool-season turfgrasses - for soil water NO3 concentration and leaching potential. Crop Sci 37 (2):526-534 Livesley SJ, Dougherty BJ, Smith AJ, Navaud D, Wylie LJ, Arndt SK (2010) Soil- atmosphere exchange of carbon dioxide, methane and nitrous oxide in urban garden systems: Impact of irrigation, fertiliser and mulch. Urban Ecosyst 13 (3):273-293 269 Livingston M, Shaw WW, Harris LK (2003) A model for assessing wildlife habitats in urban landscapes of eastern Pima County, Arizona (USA). Landsc Urban Plan 64:131-144 Loram A, Thompson K, Warren PH, Gaston KJ (2008a) Urban doemestic gardens (XII): The richness and composition of the flora in five UK cities. J Veg Sci 19 (3):321- 330 Loram A, Tratalos J, Warren PH, Gaston KJ (2007) Urban domestic gardens (X): The extent & structure of the resource in five major cities. Landsc Ecol 22 (4):601-615 Loram A, Warren PH, Gaston KJ (2008b) Urban domestic gardens (XIV): The characteristics of gardens in five cities. Environ Manag 42 (3):361-376 Loss SR, Ruiz MO, Brawn JD (2009) Relationships between avian diversity, neighborhood age, income, and environmental characteristics of an urban landscape. Biol Conserv 142 (11):2578-2585 Luck GW, Smallbone LT, O'Brien R (2009) Socio-economics and vegetation change in urban ecosystems: Patterns in space and time. Ecosyst 12 (4):604-620 MacArthur RH, Wilson EO (1967) The theory of island biogeography. Princeton University Press, Princeton, NJ Marco A, Dutoit T, Deschamps-Cottin M, Mauffrey JF, Vennetier M, Bertaudiere- Montes V (2008) Gardens in urbanizing rural areas reveal an unexpected floral diversity related to housing density. Comptes Rendus Biologies 331 (6):452-465 Martin CA (2001) Landscape water use in Phoenix, Arizona. Desert Plants 17:26-31 Martin CA (2008) Landscape sustainability in a Sonoran desert city. CATE 1 (2) Martin CA, Busse K, Yabiku S (2007) North Desert Village: The effect of landscape manipulation on microclimate and its relation to human landscape preference. Hortsci 42 (4):853-854 Martin CA, Peterson KA, Stabler LB (2003) Residential landscaping in Phoenix, Arizona, USA: Practices and preferences relative to covenants, codes and restrictions. J Arboric 29 (1):9-17 Martin CA, Warren PS, Kinzig AP (2004) Neighborhood socioeconomic status is a useful predictor of perennial landscape vegetation in residential neighborhoods and embedded small parks of Phoenix, AZ. Landsc Urban Plan 69 (4):355-368 270 Marzluff JM, Ewing K (2001) Restoration of fragmented landscapes for the conservation of birds: A general framework and specific recommendations for urbanizing landscapes. Restor Ecol 9 (3):280-292 Mathieu R, Freeman C, Aryal J (2007) Mapping private gardens in urban areas using object-oriented techniques and very high-resolution satellite imagery. Landsc Urban Plan 81 (3):179-192 Matson PA, Naylor R, Ortiz-Monasterio I (1998) Integration of environmental, agronomic, and economic aspects of fertilizer management. Sci 280 (5360):112- 115 Matsuoka RH, Kaplan R (2008) People needs in the urban landscape: Analysis of landscape and urban planning contributions. Landsc Urban Plan 84 (1):7-19 Mayer PW, DeOreo WB, et al. (1999) Residential end uses of water. American Water Works Association Research Foundation, Denver, CO McGuirk PM, Dowling R (2007) Understanding master-planned estates in Australian cities: A framework for research. Urban Policy Res 25 (1):21-38 McIntyre NE, Hostetler ME (2001) Effects of urban land use on pollinator (Hymenoptera: Apoidea) communities in a desert metropolis. Basic Appl Ecol 2 (3):209-218 McKenzie E (1994) Privatopia: Homeowner associations and the rise of residential private government. Yale University Press, New Haven, CT McKinney ML (2006) Urbanization as a major cause of biotic homogenization. Biol Conserv 127 (3):247-260 MEA (2005) Ecosystems and human well-being: Synthesis. Millenium Ecosystem Assessment. Island Press, Washington, DC Mennis J (2006) Socioeconomic-vegetation relationships in urban, residential land: The case of Denver, Colorado. Photogramm Eng Remote Sens 72 (8):911-921 Milesi C, Running SW, Elvidge CD, Dietz JB, Tuttle BT, Nemani RR (2005) Mapping and modeling the biogeochemical cycling of turf grasses in the United States. Environ Manag 36 (3):426-438 Mittelbach GG, Steiner CF, Scheiner SM, Gross KL, Reynolds HL, Waide RB, Willig MR, Dodson SI, Gough L (2001) What is the observed relationship between species richness and productivity? Ecol 82 (9):2381-2396 271 Mohamed R (2009) Why do residential developers prefer large exurban lots? Infrastructure costs and exurban development. Environ Plan B-Plan Des 36 (1):12-29 Mustafa D, Smucker TA, Ginn F, Johns R, Connely S (2010) Xeriscape people and the cultural politics of turfgrass transformation. Environ Plan D-Soc Space 28 (4):600-617 Nassauer JI, Wang ZF, Dayrell E (2009) What will the neighbors think? Cultural norms and ecological design. Landsc Urban Plan 92 (3-4):282-292 NGA (2004) Environmental lawn and garden survey. National Gardening Association, South Burlington, VT NGA (2007) Residential lawn and landscape services and the value of landscaping. National Gardening Association, South Burlington, Vermont Nielson L, Smith CL (2005) Influences on residential yard care and water quality: Tualatin watershed, Oregon. J Am Water Resour Assoc 41 (1):93-106 Nowak DJ, Rowntree RA, McPherson EG, Sisinni SM, Kerkmann ER, Stevens JC (1996) Measuring and analyzing urban tree cover. Landsc Urban Plan 36 (1):49-57 Oki L, Greco S, Gan J, Yates M, Haver D (2007) Characterizing the runoff from single- family residential landscapes. Hortsci 42 (4):911-911 Osmond DL, Hardy DH (2004) Characterization of turf practices in five North Carolina communities. J Environ Qual 33 (2):565-575 Osmond DL, Platt J (2000) Characterization of suburban nitrogen fertilizer and water use on residential turf in Cary, North Carolina. HortTechnol 10 (2):320-325 Perry T, Nawaz R (2008) An investigation into the extent and impacts of hard surfacing of domestic gardens in an area of Leeds, United Kingdom. Landsc Urban Plan 86 (1):1-13 Peters EB, McFadden JP (2010) Influence of seasonality and vegetation type on suburban microclimates. Urban Ecosyst 13:443-460. Petrovic AM (1990) The fate of nitrogenous fertilizers applied to turfgrass. J Environ Qual 19 (1):1-14 Pickett STA, Burch WR, Dalton SE, Foresman TW, Grove JM, Rowntree R (1997) A conceptual framework for the study of human ecosystems in urban areas. Urban Ecosyst 1 (4):185-199 272 Pickett STA, Cadenasso ML (2009) Altered resources, disturbance, and heterogeneity: A framework for comparing urban and non-urban soils. Urban Ecosyst 12 (1):23-44 Pickett STA, Cadenasso ML, Grove JM, Groffman PM, Band LE, Boone CG, Burch WR, Grimmond CSB, Hom J, Jenkins JC, Law NL, Nilon CH, Pouyat RV, Szlavecz K, Warren PS, Wilson MA (2008) Beyond urban legends: An emerging framework of urban ecology, as illustrated by the Baltimore Ecosystem Study. Biosci 58 (2):139-150 Polebitski AS, Palmer RN (2010) Seasonal residential water demand forecasting for census tracts. J Water Resour Plan Manag-Asce 136 (1):27-36 Pouyat RV, Groffman PM, Yesilonis ID, Hernandez L (2002) Soil carbon pools and fluxes in urban ecosystems. Environ Pollut S107-S118 Pouyat RV, Yesilonis ID, Golubiewski NE (2009) A comparison of soil organic carbon stocks between residential turfgrass and native soil. Urban Ecosyst 12 (1):45-62 Pouyat RV, Yesilonis ID, Russell-Anelli J, Neerchal NK (2007) Soil chemical and physical properties that differentiate urban land-use and cover types. Soil Sci Soc Am J 71 (3):1010-1019 Prytherch DL (2002) Selling the eco-entrepreneurial city: Natural wonders and urban stratagems in Tucson, Arizona. Urban Geogr 23 (8):771-793 Qian YL, Bandaranayake W, Parton WJ, Mecham B, Harivandi MA, Mosier AR (2003) Long-term effects of clipping and nitrogen management in turfgrass on soil organic carbon and nitrogen dynamics: The CENTURY model simulation. J Environ Qual 32:1694-1700 Raciti SM, Groffman PM, Fahey TJ (2008) Nitrogen retention in urban lawns and forests. Ecol Appl 18 (7):1615-1626 Raupp MJ, Shrewsbury PM, Herms DA (2010) Ecology of herbivorous arthropods in urban landscapes. Ann Rev Entomol 55:19-38 Redman CL (1999) Human impact on ancient environments. University of Arizona Press, Tucson, AZ Redman CL, Grove JM, Kuby LH (2004) Integrating social science into the Long-Term Ecological Research (LTER) network: Social dimensions of ecological change and ecological dimensions of social change. Ecosyst 7:161-171 273 Richards NA, Mallette JR, Simpson RJ, Macie EA (1984) Residential greenspace and vegetation in a mature city - Syracuse, New York. Urban Ecol 8 (1-2):99-125 Robbins P (2007) Lawn people: How grasses, weeds, and chemicals make us who we are. Temple University Press, Philadelphia, PA Robbins P, Birkenholtz T (2003) Turfgrass revolution: Measuring the expansion of the American lawn. Land Use Policy 20 (2):181-194 Robbins P, Polderman A, Birkenholtz T (2001) Lawns and toxins: An ecology of the city. Cities 18 (6):369-380 Robbins P, Sharp JT (2003a) Producing and consuming chemicals: The moral economy of the American lawn. Econ Geogr 79 (4):425-451 Robbins P, Sharp JT (2003b) The lawn-chemical economy and its discontents. Antipode 35 (5):955-979 Romig K (2010) Community and social capital in upper-income neighborhoods: An investigation in metropolitan Phoenix. Urban Geogr 31 (8):1065-1079 Ryznar RM, Wagner TW (2001) Using remotely sensed imagery to detect urban change – viewing Detroit from space. J Am Plan Assoc 67 (3):327-336 Salvador R, Bautista-Capetillo C, Playan E (2011) Irrigiation performance in private urban landscapes: A study case in Zaragoza (Spain). Landsc Urban Plan 100:302- 311 Sandberg LA, Foster J (2005) Challenging lawn and order: Environmental discourse and lawn care reform in Canada. Environ Politics 14 (4):478-494 Scharenbroch BC, Lloyd JE, Johnson-Maynard JL (2005) Distinguishing urban soils with physical, chemical, and biological properties. Pedobiologia 49:283-296 Schwartz S (1994) Are there universal aspects in the structure and contents of human values? J Soc Issues 50 (4):19-45 Shaffer P, Wilson S, Brindle F, Baffoe-Bonnie, B (2009) Understanding permeable and impermeable surfaces: Technical report on surfacing options and cost benefit analysis. Department of Communities and Local Government, UK. www.communities.gov.uk/ publications/planningandbuilding/permeablesufacesreport. Accessed 30 Jan 2011 Shashua-Bar L, Pearlmutter D, Erell E (2009) The cooling efficiency of urban landscape strategies in a hot dry climate. Landsc Urban Plan 92 (3-4):179-186 274 Shochat E, Lerman SB, Anderies JM, Warren PS, Faeth SH, Nilon CH (2010) Invasion, competition, and biodiversity loss in urban ecosystems. Biosci 60 (3):199-208 Shochat E, Stefanov WL, Whitehouse MEA, Faeth SH (2004) Urbanization and spider diversity: Influences of human modification of habitat structure and productivity. Ecol Appl 14 (1):268-280 Shochat E, Warren PS, Faeth SH, McIntyre NE, Hope D (2006) From patterns to emerging processes in mechanistic urban ecology. Trends Ecol Evol 21 (4):186- 191 Simpson JR, McPherson EG (1998) Simulation of tree shade impacts on residential energy use for space conditioning in Sacramento. Atmos Environ 32 (1):69-74 Smetak KM, Johnsn-Maynard JL, Lloyd JE (2007) Earthworm population density and diversity in different-aged urban systems. Appl Soil Ecol 37 (1-2):161-168 Smith J, Chapman A, Eggleton P (2006a) Baseline biodiversity surveys of the soil macrofauna of London’s green spaces. Urban Ecosyst 9 (4):337-349 Smith RM, Gaston KJ, Warren PH, Thompson K (2005) Urban domestic gardens (V): Relationships between landcover composition, housing and landscape. Landsc Ecol 20 (2):235-253 Smith RM, Gaston KJ, Warren PH, Thompson K (2006b) Urban domestic gardens (VIII): Environmental correlates of invertebrate abundance. Biodivers Conserv 15 (8):2515-2545 Smith RM, Thompson K, Hodgson JG, Warren PH, Gaston KJ (2006c) Urban domestic gardens (IX): Composition and richness of the vascular plant flora, and implications for native biodiversity. Biol Conserv 129 (3):312-322 Smith RM, Warren PH, Thompson K, Gaston KJ (2006d) Urban domestic gardens (VI): Environmental correlates of invertebrate species richness. Biodivers Conserv 15 (8):2415-2438 Song K, Qin J, Gao K, Hu YH (2009) Homogenization of plant diversity in Shanghai residential areas. Yingyong Shengtai Xuebao 20 (7):1603-1607 Sovocool KA, Morgan M, Bennett D (2006) An in-depth investigation of xeriscape as a water conservation measure. J Am Water Works Assoc 98 (2):82-93 275 Sperling C, Lortie C (2010) The importance of urban backgardens on plant and invertebrate recruitment: A field microcosm experiment. Urban Ecosyst 13 (2):223-235 Spinti JE, St Hilaire R, VanLeeuwen D (2004) Balancing landscape preferences and water conservation in a desert community. HortTechnol 14 (1):72-77 St Hilaire R, Arnold MA, Wilkerson DC, Devitt DA, Hurd BH, Lesikar BJ, Lohr VI, Martin CA, McDonald GV, Morris RL, Pittenger DR, Shaw DA, Zoldoske DF (2008) Efficient water use in residential urban landscapes. Hortsci 43 (7):2081- 2092 St Hilaire R, VanLeenwen DM, Torres P (2010) Landscape preferences and water conservation choices of residents in a high desert environment. HortTech 20 (2):308-314 Stabler LB (2008) Management regimes affect woody plant productivity and water use efficiency in an urban desert ecosystem. Urban Ecosyst 11:197-211 Stabler LB, Martin CA, Brazel AJ (2005) Microclimates in a desert city were related to land use and vegetation index. Urban For Urban Green 3 (3-4):137-147 Steer CD, Grey CNB, Team AS (2006) Socio-demographic characteristics of UK families using pesticides and weed-killers. J Expo Sci Environ Epidemiol 16 (3):251-263 Steinberg T (2006) American green: The obsessive quest for the perfect lawn. 1st edition. W.W. Norton, New York Stern PC (2000) Toward a coherent theory of environmentally significant behavior. J Soc Issues 56 (3):407-424 Stewart GH, Meurk CD, Ignatieva ME, Buckley HL, Magueur A, Case BS, Hudson M, Parker M (2009) Urban biotopes of Aotearoa New Zealand (URBANZ) II: Floristics, biodiversity and conservation values of urban residential and public woodlands, Christchurch. Urban For Urban Green 8 (3):149-162 Stone JB (2004) Paving over paradise: How land use regulations promote residential imperviousness. Landsc Urban Plan 69 (1):101-113 Struger J, Fletcher T (2007) Occurrence of lawn care and agricultural pesticides in the Don River and Humber River watersheds (1998-2002). J Gt Lakes Res 33 (4):887-905 276 Su WZ, Gu CL, Yang GS (2010) Assessing the impact of land use/land cover on urban heat island pattern in Nanjing City, China. J Urban Plan Dev-ASCE 136 (4):365- 372 Syme GJ, Shao QX, Po M, Campbell E (2004) Predicting and understanding home garden water use. Landsc Urban Plan 68 (1):121-128 Templeton SR, Yoo SJ, Zilberman D (1999) An economic analysis of yard care and synthetic chemical use: The case of San Francisco. Environ Resour Econ 14 (3):385-397 Templeton SR, Zilberman D, Yoo SJ, Dabalen AL (2008) Household use of agricultural chemicals for soil-pest management and own labor for yard work. Environ Resour Econ 40 (1):91-108 Thompson K, Austin KC, Smith RM, Warren PH, Angold PG, Gaston KJ (2003) Urban domestic gardens (I): Putting small-scale plant diversity in context. J Veg Sci 14 (1):71-78 Thompson K, Hodgson JG, Smith RM, Warren PH, Gaston KJ (2004) Urban domestic gardens (III): Composition and diversity of lawn floras. J Veg Sci 15 (3):373-378 Tilman D, Cassman KG, Matson PA, Naylor R, Polasky S (2002) Agricultural sustainability and intensive production practices. Nature 418 (6898):671-677 Townsend-Small A, Czimczik CI (2010) Carbon sequestration and greenhouse gas emissions in urban turf. Geophys Res Lett 37: L02707 Tratalos J, Fuller RA, Evans KL, Davies RG, Newson SE, Greenwood JJD, Gaston KJ (2007a) Bird densities are associated with household densities. Glob Change Biol 13 (8):1685-1695 Tratalos J, Fuller RA, Warren PH, Davies RG, Gaston KJ (2007b) Urban form, biodiversity potential and ecosystem services. Landsc Urban Plan 83 (4):308-317 Troy AR, Grove JM, O'Neil-Dunne JPM, Pickett STA, Cadenasso ML (2007) Predicting opportunities for greening and patterns of vegetation on private urban lands. Environ Manag 40 (3):394-412 Troy P, Holloway D (2004) The use of residential water consumption as an urban planning tool: A pilot study in Adelaide. J Environ Plan Manag 47 (1):97-114 Trudgill S, Jeffery A, Parker J (2010) Climate change and the resilience of the domestic lawn. Appl Geogr 30 (1):177-190 277 Tzoulas K, Korpela K, Venn S, Yli-Pelkonen V, Kazmierczak A, Niemela J, James P (2007) Promoting ecosystem and human health in urban areas using green infrastructure: A literature review. Landsc Urban Plan 81 (3):167-178 UN (2010) World urbanization prospects: The 2009 revision. United Nations, NY Uslu A, Gokce S (2010) Social interaction in urban transformation areas and the characteristics of urban outdoor spaces: A case study from Turkey. Afr J Agric Res 5 (20):2801-2810 Waide RB, Willig MR, Steiner CF, Mittelbach G, Gough L, Dodson SI, Juday GP, Parmenter R (1999) The relationship between productivity and species richness. Ann Rev Ecol Evol Syst 30:257-300 Wentz EA, Gober P (2007) Determinants of small-area water consumption for the city of Phoenix, Arizona. Water Resour Manage 21 (11):1849-1863 Whittaker D, Vaske JJ, Manfredo MJ (2006) Specificity and the cognitive hierarchy: Value orientations and the acceptability of urban wildlife management actions. Soc Nat Resour 19 (6):515-530 Wu J, Jones KB, Li H, Loucks OL (2006) Scaling and Uncertainty Analysis in Ecology: Methods and Applications. Springer, The Netherlands. Yabiku ST, Casagrande DG, Farley-Metzger E (2008) Preferences for landscape choice in a Southwestern desert city. Environ Behav 40 (3):382-400 Yahner RH (2001) Butterfly communities in residential landscapes of central Pennsylvania. Northeast Nat 8 (1):113-118 Zheng B, Zhang YQ, Chen JQ (2011) Preference to home landscape: Wildness or neatness? Landsc Urban Plan 99 (1):1-8 Zhou WQ, Troy A, Grove JM, Jenkins JC (2009) Can money buy green? Demographic and socioeconomic predictors of lawn-care expenditures and lawn greenness in urban residential areas. Soc Nat Resour 22 (8):744-760 Zhou WQ, Troy A, Grove M (2008) Modeling residential lawn fertilization practices: Integrating high resolution remote sensing with socioeconomic data. Environ Manag 41 (5):742-752 Zmyslony J, Gagnon D (1998) Residential management of urban front yard landscape: A random process? Landsc Urban Plan 40 (4):295-307 278 Zmyslony J, Gagnon D (2000) Path analysis of spatial predictors of front yard landscape in an anthropogenic environment. Landsc Ecol 15 (4):357-371 CHAPTER 3 Aber, John D., W. McDowell, K. Nadelhoffer, A. Magill, G. Berntson, M. Kamakea, S. McNulty, W. Currie, L. Rustad, and Ivan Fernandez. 1998. “Nitrogen Saturation in Temperate Forests: Hypotheses Revisited.” Bioscience 48 (11): 921–34. Afif, Charbel, Alain L. Dutot, Corinne Jambert, Maher Abboud, Jocelyne Adjizian- Gérard, Wehbeh Farah, Pascal E. Perros, and Toufic Rizk. 2009. “Statistical Approach for the Characterization of NO2 Concentrations in Beirut.” Air Quality, Atmosphere & Health 2 (2): 57–67. doi:10.1007/s11869-009-0034-2. Alonso, R., A. Bytnerowicz, and W. I. Boarman. 2005. “Atmospheric Dry Deposition in the Vicinity of the Salton Sea, California—I: Air Pollution and Deposition in a Desert Environment.” Atmospheric Environment 39 (26): 4671–79. doi:10.1016/j.atmosenv.2005.04.016. Asman, Willem a. H., Mark A. Sutton, and Jan K. Schjørring. 1998. “Ammonia: Emission, Atmospheric Transport and Deposition.” New Phytologist 139 (1): 27– 48. doi:10.1046/j.1469-8137.1998.00180.x. Báez, S., J. Fargione, D. I. Moore, S. L. Collins, and J. R. Gosz. 2007. “Atmospheric Nitrogen Deposition in the Northern Chihuahuan Desert: Temporal Trends and Potential Consequences.” Journal of Arid Environments 68 (4): 640–51. doi:10.1016/j.jaridenv.2006.06.011. Baker, Lawrence A., Diane Hope, Ying Xu, Jennifer Edmonds, and Lisa Lauver. 2001. “Nitrogen Balance for the Central Arizona–Phoenix (CAP) Ecosystem.” Ecosystems 4 (6): 582–602. doi:10.1007/s10021-001-0031-2. Baron, Jill S. 2006. “Hindcasting Nitrogen Deposition To Determine An Ecological Critical Load.” Ecological Applications 16 (2): 433–39. doi:10.1890/1051- 0761(2006)016[0433:HNDTDA]2.0.CO;2. Bettez, Neil D., and Peter M. Groffman. 2013. “Nitrogen Deposition in and near an Urban Ecosystem.” Environmental Science & Technology 47 (11): 6047–51. doi:10.1021/es400664b. Bobbink, Roland, K. Hicks, J. Galloway, T. Spranger, R. Alkemade, M. Ashmore, M. 279 Bustamante, S. Cinderby, E. Davidson, and F. Dentener. 2010. “Global Assessment of Nitrogen Deposition Effects on Terrestrial Plant Diversity: A Synthesis.” Ecological Applications 20 (1): 30–59. Brazel, A., N. Selover, R. Vose, and G. Heisler. 2000. “The Tale of Two Climates - Baltimore and Phoenix Urban LTER Sites.” Climate Research 15 (2): 123–35. doi:10.3354/cr015123. Brooks, Matthew L. 2003. “Effects of Increased Soil Nitrogen on the Dominance of Alien Annual Plants in the Mojave Desert.” Journal of Applied Ecology 40 (2): 344–53. doi:10.1046/j.1365-2664.2003.00789.x. Bytnerowicz, Andrzej, Kenji Omasa, and Elena Paoletti. 2007. “Integrated Effects of Air Pollution and Climate Change on Forests: A Northern Hemisphere Perspective.” Environmental Pollution, Air Pollution and Climate Change: A Global Overview of the Effects on Forest Vegetation, 147 (3): 438–45. doi:10.1016/j.envpol.2006.08.028. Bytnerowicz, Andrzej, Maria Jose Sanz, Michael J. Arbaugh, Pamela E. Padgett, David P. Jones, and Antonio Davila. 2005. “Passive Sampler for Monitoring Ambient Nitric Acid (HNO3) and Nitrous Acid (HNO2) Concentrations.” Atmospheric Environment 39 (14): 2655–60. doi:10.1016/j.atmosenv.2005.01.018. Cisneros, Ricardo, Andrzej Bytnerowicz, Donald Schweizer, Sharon Zhong, Samuel Traina, and Deborah H. Bennett. 2010. “Ozone, Nitric Acid, and Ammonia Air Pollution Is Unhealthy for People and Ecosystems in Southern Sierra Nevada, California.” Environmental Pollution 158 (10): 3261–71. doi:10.1016/j.envpol.2010.07.025. Clark, Christopher M., and David Tilman. 2008. “Loss of Plant Species after Chronic Low-Level Nitrogen Deposition to Prairie Grasslands.” Nature 451 (7179): 712– 15. doi:10.1038/nature06503. Collins, Scott L., Robert L. Sinsabaugh, Chelsea Crenshaw, Laura Green, Andrea Porras- Alfaro, Martina Stursova, and Lydia H. Zeglin. 2008. “Pulse Dynamics and Microbial Processes in Aridland Ecosystems.” Journal of Ecology 96 (3): 413–20. Delon, C., C. Galy-Lacaux, M. Adon, C. Liousse, D. Serça, B. Diop, and A. Akpo. 2012. “Nitrogen Compounds Emission and Deposition in West African Ecosystems: Comparison between Wet and Dry Savanna.” Biogeosciences 9 (1): 385–402. doi:10.5194/bg-9-385-2012. Dentener, F., J. Drevet, J. F. Lamarque, I. Bey, B. Eickhout, A. M. Fiore, D. Hauglustaine, et al. 2006. “Nitrogen and Sulfur Deposition on Regional and 280 Global Scales: A Multimodel Evaluation.” Global Biogeochemical Cycles 20 (4): GB4003. doi:10.1029/2005GB002672. Ellis, R. A., D. J. Jacob, M. P. Sulprizio, L. Zhang, C. D. Holmes, B. A. Schichtel, T. Blett, E. Porter, L. H. Pardo, and J. A. Lynch. 2013. “Present and Future Nitrogen Deposition to National Parks in the United States: Critical Load Exceedances.” Atmos. Chem. Phys. 13 (17): 9083–95. doi:10.5194/acp-13-9083-2013. Fenn, Mark E., E. B. Allen, S. B. Weiss, S. Jovan, L. H. Geiser, G. S. Tonnesen, R. F. Johnson, et al. 2010. “Nitrogen Critical Loads and Management Alternatives for N-Impacted Ecosystems in California.” Journal of Environmental Management 91 (12): 2404–23. doi:10.1016/j.jenvman.2010.07.034. Fenn, Mark E., JILL S. Baron, EDITH B. Allen, HEATHER M. Rueth, KOREN R. Nydick, LINDA Geiser, WILLIAM D. Bowman, et al. 2003. “Ecological Effects of Nitrogen Deposition in the Western United States.” BioScience 53 (4): 404–20. doi:10.1641/0006-3568(2003)053[0404:EEONDI]2.0.CO;2. Fenn, Mark E., Richard Haeuber, Gail S. Tonnesen, Jill S. Baron, Susanne Grossman- Clarke, Diane Hope, Daniel A. Jaffe, Scott Copeland, Linda Geiser, and Heather M. Rueth. 2003. “Nitrogen Emissions, Deposition, and Monitoring in the Western United States.” BioScience 53 (4): 391–403. Fenn, Mark E., and Mark A. Poth. 2004. “Monitoring Nitrogen Deposition in Throughfall Using Ion Exchange Resin Columns.” Journal of Environment Quality 33 (6): 2007. doi:10.2134/jeq2004.2007. Fenn, Mark E., Christopher S. Ross, Susan L. Schilling, William D. Baccus, Michael A. Larrabee, and Rebecca A. Lofgren. 2013. “Atmospheric Deposition of Nitrogen and Sulfur and Preferential Canopy Consumption of Nitrate in Forests of the Pacific Northwest, USA.” Forest Ecology and Management 302 (August): 240– 53. doi:10.1016/j.foreco.2013.03.042. Galloway, James N., FRANK J. Dentener, DOUGLAS G. Capone, ELISABETH W. Boyer, ROBERT W. Howarth, SYBIL P. Seitzinger, GREGORY P. Asner, C. C. Cleveland, P. A. Green, and E. A. Holland. 2004. “Nitrogen Cycles: Past, Present, and Future.” Biogeochemistry 70 (2): 153–226. Golden, Heather E., Elizabeth W. Boyer, Michael G. Brown, Emily M. Elliott, and Don Koo Lee. 2008. “Simple Approaches for Measuring Dry Atmospheric Nitrogen Deposition to Watersheds.” Water Resources Research 44 (4): W00D02. doi:10.1029/2008WR006952. Grimm, Nancy B., and Charles L. Redman. 2004. “Approaches to the Study of Urban 281 Ecosystems: The Case of Central Arizona—Phoenix.” Urban Ecosystems 7 (3): 199–213. doi:10.1023/B:UECO.0000044036.59953.a1. Hall, Sharon J., Ryan A. Sponseller, Nancy B. Grimm, David Huber, Jason P. Kaye, Christopher Clark, and Scott L. Collins. 2011. “Ecosystem Response to Nutrient Enrichment across an Urban Airshed in the Sonoran Desert.” Ecological Applications 21 (3): 640–60. doi:10.1890/10-0758.1. Hanson, Paul J., and Steven E. Lindberg. 1991. “Dry Deposition of Reactive Nitrogen Compounds: A Review of Leaf, Canopy and Non-Foliar Measurements.” Atmospheric Environment. Part A. General Topics 25 (8): 1615–34. doi:10.1016/0960-1686(91)90020-8. Holland, Elisabeth A., Bobby H. Braswell, James Sulzman, and Jean-Francois Lamarque. 2005. “NITROGEN DEPOSITION ONTO THE UNITED STATES AND WESTERN EUROPE: SYNTHESIS OF OBSERVATIONS AND MODELS.” Ecological Applications 15 (1): 38–57. doi:10.1890/03-5162. Koutrakis, P., C. Sioutas, S. T. Ferguson, J. M. Wolfson, James D. Mulik, and Robert M. Burton. 1993. “Development and Evaluation of a Glass Honeycomb Denuder/filter Pack System to Collect Atmospheric Gases and Particles.” Environmental Science & Technology 27 (12): 2497–2501. doi:10.1021/es00048a029. Krupa, S. V, and A. H Legge. 2000. “Passive Sampling of Ambient, Gaseous Air Pollutants: An Assessment from an Ecological Perspective.” Environmental Pollution 107 (1): 31–45. doi:10.1016/S0269-7491(99)00154-2. Lee, Sang-Mi, H. J. S. Fernando, and S. Grossman-Clarke. 2007. “MM5-SMOKE- CMAQ as a Modeling Tool for 8-H Ozone Regulatory Enforcement: Application to the State of Arizona.” Environmental Modeling & Assessment 12 (1): 63–74. doi:10.1007/s10666-006-9053-7. Li, Kaihui, Xuejun Liu, Wei Song, Yunhua Chang, Yukun Hu, and Changyan Tian. 2013. “Atmospheric Nitrogen Deposition at Two Sites in an Arid Environment of Central Asia.” PLoS ONE 8 (6): e67018. doi:10.1371/journal.pone.0067018. Lohse, Kathleen A., Diane Hope, Ryan Sponseller, Jonathan O. Allen, and Nancy B. Grimm. 2008. “Atmospheric Deposition of Carbon and Nutrients across an Arid Metropolitan Area.” Science of The Total Environment 402 (1): 95–105. doi:10.1016/j.scitotenv.2008.04.044. Lovett, Gary M. 1994. “Atmospheric Deposition of Nutrients and Pollutants in North America: An Ecological Perspective.” Ecological Applications 4 (4): 629–50. 282 ———. 2013. “Critical Issues for Critical Loads.” Proceedings of the National Academy of Sciences 110 (3): 808–9. doi:10.1073/pnas.1219007110. Lovett, Gary M., Martin M. Traynor, Richard V. Pouyat, Margaret M. Carreiro, Wei- Xing Zhu, and James W. Baxter. 2000. “Atmospheric Deposition to Oak Forests along an Urban-Rural Gradient.” Environmental Science & Technology 34 (20): 4294–4300. Moodley, Kandasamy G., Shalini Singh, and Suendran Govender. 2011. “Passive Monitoring of Nitrogen Dioxide in Urban Air: A Case Study of Durban Metropolis, South Africa.” Journal of Environmental Management 92 (9): 2145– 50. doi:10.1016/j.jenvman.2011.03.040. Nunnermacker, L. J., J. Weinstein-Lloyd, L. Kleinman, P. H. Daum, Y. N. Lee, S. R. Springston, P. Klotz, L. Newman, G. Neuroth, and P. Hyde. 2004. “Ground- Based and Aircraft Measurements of Trace Gases in Phoenix, Arizona (1998).” Atmospheric Environment 38 (29): 4941–56. doi:10.1016/j.atmosenv.2004.04.033. Pan, Y. P., Y. S. Wang, G. Q. Tang, and D. Wu. 2012. “Wet and Dry Deposition of Atmospheric Nitrogen at Ten Sites in Northern China.” Atmos. Chem. Phys. 12 (14): 6515–35. doi:10.5194/acp-12-6515-2012. Pardo, Linda H., Mark E. Fenn, Christine L. Goodale, Linda H. Geiser, Charles T. Driscoll, Edith B. Allen, Jill S. Baron, et al. 2011. “Effects of Nitrogen Deposition and Empirical Nitrogen Critical Loads for Ecoregions of the United States.” Ecological Applications 21 (8): 3049–82. doi:10.1890/10-2341.1. Phoenix, Gareth K., W. Kevin Hicks, Steve Cinderby, Johan C. I. Kuylenstierna, William D. Stock, Frank J. Dentener, Ken E. Giller, et al. 2006. “Atmospheric Nitrogen Deposition in World Biodiversity Hotspots: The Need for a Greater Global Perspective in Assessing N Deposition Impacts.” Global Change Biology 12 (3): 470–76. doi:10.1111/j.1365-2486.2006.01104.x. Rao, Leela E., and Edith B. Allen. 2010. “Combined Effects of Precipitation and Nitrogen Deposition on Native and Invasive Winter Annual Production in California Deserts.” Oecologia 162 (4): 1035–46. doi:10.1007/s00442-009-1516- 5. Rao, Leela E., Edith B. Allen, and Thomas Meixner. 2010. “Risk-Based Determination of Critical Nitrogen Deposition Loads for Fire Spread in Southern California Deserts.” Ecological Applications 20 (5): 1320–35. doi:10.1890/09-0398.1. Rao, Leela E., D. R. Parker, A. Bytnerowicz, and E. B. Allen. 2009. “Nitrogen 283 Mineralization across an Atmospheric Nitrogen Deposition Gradient in Southern California Deserts.” Journal of Arid Environments 73 (10): 920–30. doi:10.1016/j.jaridenv.2009.04.007. Rao, Preeti, Lucy R. Hutyra, Steve M. Raciti, and Adrien C. Finzi. 2013. “Field and Remotely Sensed Measures of Soil and Vegetation Carbon and Nitrogen across an Urbanization Gradient in the Boston Metropolitan Area.” Urban Ecosystems 16 (3): 593–616. doi:10.1007/s11252-013-0291-6. Roadman, M. J, J. R Scudlark, J. J Meisinger, and W. J Ullman. 2003. “Validation of Ogawa Passive Samplers for the Determination of Gaseous Ammonia Concentrations in Agricultural Settings.” Atmospheric Environment 37 (17): 2317–25. doi:10.1016/S1352-2310(03)00163-8. Salem, Alaa A., Ahmed A. Soliman, and Ismail A. El-Haty. 2009. “Determination of Nitrogen Dioxide, Sulfur Dioxide, Ozone, and Ammonia in Ambient Air Using the Passive Sampling Method Associated with Ion Chromatographic and Potentiometric Analyses.” Air Quality, Atmosphere & Health 2 (3): 133–45. doi:10.1007/s11869-009-0040-4. Sather, Mark E., Johnson Mathew, Nghia Nguyen, John Lay, George Golod, Robert Vet, Joseph Cotie, et al. 2008. “Baseline Ambient Gaseous Ammonia Concentrations in the Four Corners Area and Eastern Oklahoma, USA.” Journal of Environmental Monitoring 10 (11): 1319. doi:10.1039/b807984f. Sather, Mark E., E. Terrence Slonecker, Johnson Mathew, Hunter Daughtrey, and Dennis D. Williams. 2007. “Evaluation of Ogawa Passive Sampling Devices as an Alternative Measurement Method for the Nitrogen Dioxide Annual Standard in El Paso, Texas.” Environmental Monitoring and Assessment 124 (1-3): 211–21. doi:10.1007/s10661-006-9219-4. Schwede, Donna, Leiming Zhang, Robert Vet, and Gary Lear. 2011. “An Intercomparison of the Deposition Models Used in the CASTNET and CAPMoN Networks.” Atmospheric Environment 45 (6): 1337–46. doi:10.1016/j.atmosenv.2010.11.050. Simkin, Samuel M., David N. Lewis, Kathleen C. Weathers, Gary M. Lovett, and Kirsten Schwarz. 2004. “Determination of Sulfate, Nitrate, and Chloride in Throughfall Using Ion-Exchange Resins.” Water, Air, and Soil Pollution 153 (1-4): 343–54. doi:10.1023/B:WATE.0000019958.59277.ed. Sponseller, Ryan A., Sharon J. Hall, David P. Huber, Nancy B. Grimm, Jason P. Kaye, Christopher M. Clark, and Scott L. Collins. 2012. “Variation in Monsoon Precipitation Drives Spatial and Temporal Patterns of Larrea Tridentata Growth in the Sonoran Desert.” Functional Ecology 26 (3): 750–58. doi:10.1111/j.1365- 284 2435.2012.01979.x. Vitousek, Peter M., John D. Aber, Robert W. Howarth, Gene E. Likens, Pamela A. Matson, David W. Schindler, William H. Schlesinger, and David G. Tilman. 1997. “HUMAN ALTERATION OF THE GLOBAL NITROGEN CYCLE: SOURCES AND CONSEQUENCES.” Ecological Applications 7 (3): 737–50. doi:10.1890/1051-0761(1997)007[0737:HAOTGN]2.0.CO;2. Wang, Ge, and Martin Ostoja-Starzewski. 2004. “Influence of Topography on the Phoenix CO2 Dome: A Computational Study.” Atmospheric Science Letters 5 (5): 103–7. doi:10.1002/asl.67. Watson, John G., Judith C. Chow, Fred W. Lurmann, and Stefan P. Musarra. 1994. “Ammonium Nitrate, Nitric Acid, and Ammonia Equilibrium in Wintertime Phoenix, Arizona.” Air & Waste 44 (4): 405–12. doi:10.1080/1073161X.1994.10467262. Weathers, Kathleen C., G. M. Lovett, G. E. Likens, and R. Lathrop. 2000. “The Effect of Landscape Features on Deposition to Hunter Mountain, Catskill Mountains, New York.” Ecological Applications 10 (2): 528–40. Weathers, Kathleen C., Samuel M. Simkin, Gary M. Lovett, and Steven E. Lindberg. 2006. “Empirical Modeling of Atmospheric Deposition in Mountainous Landscapes.” Ecological Applications 16 (4): 1590–1607. Wesely, M. L, and B. B Hicks. 2000. “A Review of the Current Status of Knowledge on Dry Deposition.” Atmospheric Environment 34 (12–14): 2261–82. doi:10.1016/S1352-2310(99)00467-7. Zbieranowski, Antoni L., and Julian Aherne. 2012. “Ambient Concentrations of Atmospheric Ammonia, Nitrogen Dioxide and Nitric Acid across a Rural–urban– agricultural Transect in Southern Ontario, Canada.” Atmospheric Environment 62 (December): 481–91. doi:10.1016/j.atmosenv.2012.08.040. Zhang, L., J. R. Brook, and R. Vet. 2003. “A Revised Parameterization for Gaseous Dry Deposition in Air-Quality Models.” Atmospheric Chemistry and Physics 3 (6): 2067–82. Zhang, L., R. Vet, J. M. O’Brien, C. Mihele, Z. Liang, and A. Wiebe. 2009. “Dry Deposition of Individual Nitrogen Species at Eight Canadian Rural Sites.” Journal of Geophysical Research: Atmospheres 114 (D2): n/a–n/a. doi:10.1029/2008JD010640. 285 CHAPTER 4 Aber, John D., W. McDowell, K. Nadelhoffer, A. Magill, G. Berntson, M. Kamakea, S. McNulty, W. Currie, L. Rustad, and Ivan Fernandez. 1998. “Nitrogen Saturation in Temperate Forests: Hypotheses Revisited.” Bioscience 48 (11): 921–34. Ainsworth, Elizabeth A., Craig R. Yendrek, Stephen Sitch, William J. Collins, and Lisa D. Emberson. 2012. “The Effects of Tropospheric Ozone on Net Primary Productivity and Implications for Climate Change*.” Annual Review of Plant Biology 63 (1): 637–61. doi:10.1146/annurev-arplant-042110-103829. Akimoto, Hajime. 2003. “Global Air Quality and Pollution.” Science 302 (5651): 1716– 19. doi:10.1126/science.1092666. Alonso, R., A. Bytnerowicz, and W. I. Boarman. 2005. “Atmospheric Dry Deposition in the Vicinity of the Salton Sea, California—I: Air Pollution and Deposition in a Desert Environment.” Atmospheric Environment 39 (26): 4671–79. doi:10.1016/j.atmosenv.2005.04.016. Atkinson-Palombo, Carol M., James A. Miller, and Robert C. Balling Jr. 2006. “Quantifying the Ozone ‘weekend Effect’ at Various Locations in Phoenix, Arizona.” Atmospheric Environment 40 (39): 7644–58. doi:10.1016/j.atmosenv.2006.05.023. Báez, S., J. Fargione, D. I. Moore, S. L. Collins, and J. R. Gosz. 2007. “Atmospheric Nitrogen Deposition in the Northern Chihuahuan Desert: Temporal Trends and Potential Consequences.” Journal of Arid Environments 68 (4): 640–51. doi:10.1016/j.jaridenv.2006.06.011. Blanchard, Charles L., Shelley Tanenbaum, and Douglas R. Lawson. 2008. “Differences between Weekday and Weekend Air Pollutant Levels in Atlanta; Baltimore; Chicago; Dallas–Fort Worth; Denver; Houston; New York; Phoenix; Washington, DC; and Surrounding Areas.” Journal of the Air & Waste Management Association 58 (12): 1598–1615. doi:10.3155/1047-3289.58.12.1598. Bobbink, Roland, K. Hicks, J. Galloway, T. Spranger, R. Alkemade, M. Ashmore, M. Bustamante, S. Cinderby, E. Davidson, and F. Dentener. 2010. “Global Assessment of Nitrogen Deposition Effects on Terrestrial Plant Diversity: A Synthesis.” Ecological Applications 20 (1): 30–59. Brooks, Matthew L. 2003. “Effects of Increased Soil Nitrogen on the Dominance of Alien Annual Plants in the Mojave Desert.” Journal of Applied Ecology 40 (2): 344–53. doi:10.1046/j.1365-2664.2003.00789.x. 286 Bytnerowicz, Andrzej, Maria Jose Sanz, Michael J. Arbaugh, Pamela E. Padgett, David P. Jones, and Antonio Davila. 2005. “Passive Sampler for Monitoring Ambient Nitric Acid (HNO3) and Nitrous Acid (HNO2) Concentrations.” Atmospheric Environment 39 (14): 2655–60. doi:10.1016/j.atmosenv.2005.01.018. Cisneros, Ricardo, Andrzej Bytnerowicz, Donald Schweizer, Sharon Zhong, Samuel Traina, and Deborah H. Bennett. 2010. “Ozone, Nitric Acid, and Ammonia Air Pollution Is Unhealthy for People and Ecosystems in Southern Sierra Nevada, California.” Environmental Pollution 158 (10): 3261–71. doi:10.1016/j.envpol.2010.07.025. Clark, Christopher M., and David Tilman. 2008. “Loss of Plant Species after Chronic Low-Level Nitrogen Deposition to Prairie Grasslands.” Nature 451 (7179): 712– 15. doi:10.1038/nature06503. Coutts, Andrew M., Jason Beringer, and Nigel J. Tapper. 2007. “Characteristics Influencing the Variability of Urban CO2 Fluxes in Melbourne, Australia.” Atmospheric Environment 41 (1): 51–62. doi:10.1016/j.atmosenv.2006.08.030. Day, Thomas A, Patricia Gober, Fusheng S Xiong, and Elizabeth A Wentz. 2002. “Temporal Patterns in near-Surface CO2 Concentrations over Contrasting Vegetation Types in the Phoenix Metropolitan Area.” Agricultural and Forest Meteorology 110 (3): 229–45. doi:10.1016/S0168-1923(01)00288-X. Delon, C., C. Galy-Lacaux, M. Adon, C. Liousse, D. Serça, B. Diop, and A. Akpo. 2012. “Nitrogen Compounds Emission and Deposition in West African Ecosystems: Comparison between Wet and Dry Savanna.” Biogeosciences 9 (1): 385–402. doi:10.5194/bg-9-385-2012. Dentener, F., J. Drevet, J. F. Lamarque, I. Bey, B. Eickhout, A. M. Fiore, D. Hauglustaine, et al. 2006. “Nitrogen and Sulfur Deposition on Regional and Global Scales: A Multimodel Evaluation.” Global Biogeochemical Cycles 20 (4): GB4003. doi:10.1029/2005GB002672. Duren, Riley M., and Charles E. Miller. 2012. “Measuring the Carbon Emissions of Megacities.” Nature Climate Change 2 (8): 560–62. doi:10.1038/nclimate1629. Fenn, Mark E., E. B. Allen, S. B. Weiss, S. Jovan, L. H. Geiser, G. S. Tonnesen, R. F. Johnson, et al. 2010. “Nitrogen Critical Loads and Management Alternatives for N-Impacted Ecosystems in California.” Journal of Environmental Management 91 (12): 2404–23. doi:10.1016/j.jenvman.2010.07.034. Galloway, James N., Alan R. Townsend, Jan Willem Erisman, Mateete Bekunda, Zucong Cai, John R. Freney, Luiz A. Martinelli, Sybil P. Seitzinger, and Mark A. Sutton. 287 2008. “Transformation of the Nitrogen Cycle: Recent Trends, Questions, and Potential Solutions.” Science 320 (5878): 889–92. García, Ma Ángeles, Ma Luisa Sánchez, and Isidro A. Pérez. 2012. “Differences between Carbon Dioxide Levels over Suburban and Rural Sites in Northern Spain.” Environmental Science and Pollution Research 19 (2): 432–39. doi:10.1007/s11356-011-0575-4. George, K., L. H. Ziska, J. A. Bunce, and B. Quebedeaux. 2007. “Elevated Atmospheric CO2 Concentration and Temperature across an Urban–rural Transect.” Atmospheric Environment 41 (35): 7654–65. doi:10.1016/j.atmosenv.2007.08.018. Gregg, Jillian W., Clive G. Jones, and Todd E. Dawson. 2003. “Urbanization Effects on Tree Growth in the Vicinity of New York City.” Nature 424 (6945): 183–87. doi:10.1038/nature01728. ———. 2006. “PHYSIOLOGICAL AND DEVELOPMENTAL EFFECTS OF O3 ON COTTONWOOD GROWTH IN URBAN AND RURAL SITES.” Ecological Applications 16 (6): 2368–81. doi:10.1890/1051- 0761(2006)016[2368:PADEOO]2.0.CO;2. Grimmond, C. S. B, T. S King, F. D Cropley, D. J Nowak, and C Souch. 2002. “Local- Scale Fluxes of Carbon Dioxide in Urban Environments: Methodological Challenges and Results from Chicago.” Environmental Pollution 116, Supplement 1 (March): S243–S254. doi:10.1016/S0269-7491(01)00256-1. Gurney, Kevin R., Daniel L. Mendoza, Yuyu Zhou, Marc L. Fischer, Chris C. Miller, Sarath Geethakumar, and Stephane de la Rue du Can. 2009. “High Resolution Fossil Fuel Combustion CO2 Emission Fluxes for the United States.” Environmental Science & Technology 43 (14): 5535–41. doi:10.1021/es900806c. Hall, Sharon J., Ryan A. Sponseller, Nancy B. Grimm, David Huber, Jason P. Kaye, Christopher Clark, and Scott L. Collins. 2011. “Ecosystem Response to Nutrient Enrichment across an Urban Airshed in the Sonoran Desert.” Ecological Applications 21 (3): 640–60. doi:10.1890/10-0758.1. Helfter, C., D. Famulari, G. J. Phillips, J. F. Barlow, C. R. Wood, C. S. B. Grimmond, and E. Nemitz. 2011. “Controls of Carbon Dioxide Concentrations and Fluxes above Central London.” Atmos. Chem. Phys. 11 (5): 1913–28. doi:10.5194/acp- 11-1913-2011. Hidy, George M., and William T. Pennell. 2010. “Multipollutant Air Quality Management.” Journal of the Air & Waste Management Association 60 (6): 645– 288 74. doi:10.3155/1047-3289.60.6.645. Holland, Elisabeth A., Bobby H. Braswell, James Sulzman, and Jean-Francois Lamarque. 2005. “NITROGEN DEPOSITION ONTO THE UNITED STATES AND WESTERN EUROPE: SYNTHESIS OF OBSERVATIONS AND MODELS.” Ecological Applications 15 (1): 38–57. doi:10.1890/03-5162. Idso, Craig D, Sherwood B Idso, and Robert C Balling Jr. 2001. “An Intensive Two- Week Study of an Urban CO2 Dome in Phoenix, Arizona, USA.” Atmospheric Environment 35 (6): 995–1000. doi:10.1016/S1352-2310(00)00412-X. Idso, Craig D., Sherwood B. Idso, and Robert C. Balling Jr. 1998. “The Urban Co2 Dome of Phoenix, Arizona.” Physical Geography 19 (2): 95–108. doi:10.1080/02723646.1998.10642642. Karnosky, David F., John M. Skelly, Kevin E. Percy, and Art H. Chappelka. 2007. “Perspectives Regarding 50 Years of Research on Effects of Tropospheric Ozone Air Pollution on US Forests.” Environmental Pollution 147 (3): 489–506. doi:10.1016/j.envpol.2006.08.043. Koerner, B., and J. Klopatek. 2002. “Anthropogenic and Natural CO2 Emission Sources in an Arid Urban Environment.” Environmental Pollution 116, Supplement 1 (March): S45–S51. doi:10.1016/S0269-7491(01)00246-9. Koutrakis, P., C. Sioutas, S. T. Ferguson, J. M. Wolfson, James D. Mulik, and Robert M. Burton. 1993. “Development and Evaluation of a Glass Honeycomb Denuder/filter Pack System to Collect Atmospheric Gases and Particles.” Environmental Science & Technology 27 (12): 2497–2501. doi:10.1021/es00048a029. Lee, Sang-Mi, H. J. S. Fernando, and S. Grossman-Clarke. 2007. “MM5-SMOKE- CMAQ as a Modeling Tool for 8-H Ozone Regulatory Enforcement: Application to the State of Arizona.” Environmental Modeling & Assessment 12 (1): 63–74. doi:10.1007/s10666-006-9053-7. Li, Kaihui, Xuejun Liu, Wei Song, Yunhua Chang, Yukun Hu, and Changyan Tian. 2013. “Atmospheric Nitrogen Deposition at Two Sites in an Arid Environment of Central Asia.” PLoS ONE 8 (6): e67018. doi:10.1371/journal.pone.0067018. Lohse, Kathleen A., Diane Hope, Ryan Sponseller, Jonathan O. Allen, and Nancy B. Grimm. 2008. “Atmospheric Deposition of Carbon and Nutrients across an Arid Metropolitan Area.” Science of The Total Environment 402 (1): 95–105. doi:10.1016/j.scitotenv.2008.04.044. 289 Long, Stephen P, Elizabeth A Ainsworth, Alistair Rogers, and Donald R Ort. 2004. “Rising Atmospheric Carbon Dioxide: Plants FACE the Future.” Annual Review of Plant Biology 55: 591–628. doi:10.1146/annurev.arplant.55.031903.141610. Lovett, Gary M. 1994. “Atmospheric Deposition of Nutrients and Pollutants in North America: An Ecological Perspective.” Ecological Applications 4 (4): 629–50. Lovett, Gary M., Timothy H. Tear, David C. Evers, Stuart E.G. Findlay, B. Jack Cosby, Judy K. Dunscomb, Charles T. Driscoll, and Kathleen C. Weathers. 2009. “Effects of Air Pollution on Ecosystems and Biological Diversity in the Eastern United States.” Annals of the New York Academy of Sciences 1162 (1): 99–135. doi:10.1111/j.1749-6632.2009.04153.x. Melillo, Jm, Ad Mcguire, Dw Kicklighter, B. Moore, Cj Vorosmarty, and Al Schloss. 1993. “Global Climate-Change and Terrestrial Net Primary Production.” Nature 363 (6426): 234–40. doi:10.1038/363234a0. Molina, Mario J., and Luisa T. Molina. 2004. “Megacities and Atmospheric Pollution.” Journal of the Air & Waste Management Association 54 (6): 644–80. doi:10.1080/10473289.2004.10470936. Monks, P. S., C. Granier, S. Fuzzi, A. Stohl, M. L. Williams, H. Akimoto, M. Amann, et al. 2009. “Atmospheric Composition Change – Global and Regional Air Quality.” Atmospheric Environment, ACCENT Synthesis, 43 (33): 5268–5350. doi:10.1016/j.atmosenv.2009.08.021. Nasrallah, Hassan A, Robert C Balling Jr, Shaker Mohammed Madi, and Lamya Al- Ansari. 2003. “Temporal Variations in Atmospheric CO2 Concentrations in Kuwait City, Kuwait with Comparisons to Phoenix, Arizona, USA.” Environmental Pollution 121 (2): 301–5. doi:10.1016/S0269-7491(02)00221-X. Nunnermacker, L. J., J. Weinstein-Lloyd, L. Kleinman, P. H. Daum, Y. N. Lee, S. R. Springston, P. Klotz, L. Newman, G. Neuroth, and P. Hyde. 2004. “Ground- Based and Aircraft Measurements of Trace Gases in Phoenix, Arizona (1998).” Atmospheric Environment 38 (29): 4941–56. doi:10.1016/j.atmosenv.2004.04.033. Pan, Y. P., Y. S. Wang, G. Q. Tang, and D. Wu. 2012. “Wet and Dry Deposition of Atmospheric Nitrogen at Ten Sites in Northern China.” Atmos. Chem. Phys. 12 (14): 6515–35. doi:10.5194/acp-12-6515-2012. Paoletti, Elena, and William J. Manning. 2007. “Toward a Biologically Significant and Usable Standard for Ozone That Will Also Protect Plants.” Environmental Pollution 150 (1): 85–95. doi:10.1016/j.envpol.2007.06.037. 290 Pardo, Linda H., Mark E. Fenn, Christine L. Goodale, Linda H. Geiser, Charles T. Driscoll, Edith B. Allen, Jill S. Baron, et al. 2011. “Effects of Nitrogen Deposition and Empirical Nitrogen Critical Loads for Ecoregions of the United States.” Ecological Applications 21 (8): 3049–82. doi:10.1890/10-2341.1. Parrish, David D., Hanwant B. Singh, Luisa Molina, and Sasha Madronich. 2011. “Air Quality Progress in North American Megacities: A Review.” Atmospheric Environment 45 (39): 7015–25. doi:10.1016/j.atmosenv.2011.09.039. Pataki, D. E., R. J. Alig, A. S. Fung, N. E. Golubiewski, C. A. Kennedy, E. G. Mcpherson, D. J. Nowak, R. V. Pouyat, and P. Romero Lankao. 2006. “Urban Ecosystems and the North American Carbon Cycle.” Global Change Biology 12 (11): 2092–2102. doi:10.1111/j.1365-2486.2006.01242.x. Pataki, D. E., T. Xu, Y. Q. Luo, and J. R. Ehleringer. 2007. “Inferring Biogenic and Anthropogenic Carbon Dioxide Sources across an Urban to Rural Gradient.” Oecologia 152 (2): 307–22. doi:10.1007/s00442-006-0656-0. Payne, Richard J., Nancy B. Dise, Carly J. Stevens, David J. Gowing, Cecilia Duprè, Edu Dorland, Cassandre Gaudnik, et al. 2013. “Impact of Nitrogen Deposition at the Species Level.” Proceedings of the National Academy of Sciences 110 (3): 984– 87. doi:10.1073/pnas.1214299109. Payne, Richard J., Carly J. Stevens, Nancy B. Dise, David J. Gowing, Michael G. Pilkington, Gareth K. Phoenix, Bridget A. Emmett, and Michael R. Ashmore. 2011. “Impacts of Atmospheric Pollution on the Plant Communities of British Acid Grasslands.” Environmental Pollution, Nitrogen Deposition, Critical Loads and Biodiversity, 159 (10): 2602–8. doi:10.1016/j.envpol.2011.06.009. Pope, Ronald, and Jianguo Wu. 2014. “Characterizing Air Pollution Patterns on Multiple Time Scales in Urban Areas: A Landscape Ecological Approach.” Urban Ecosystems, 1–20. Accessed March 31. doi:10.1007/s11252-014-0357-0. Rao, Leela E., Edith B. Allen, and Thomas Meixner. 2010. “Risk-Based Determination of Critical Nitrogen Deposition Loads for Fire Spread in Southern California Deserts.” Ecological Applications 20 (5): 1320–35. doi:10.1890/09-0398.1. Roadman, M. J, J. R Scudlark, J. J Meisinger, and W. J Ullman. 2003. “Validation of Ogawa Passive Samplers for the Determination of Gaseous Ammonia Concentrations in Agricultural Settings.” Atmospheric Environment 37 (17): 2317–25. doi:10.1016/S1352-2310(03)00163-8. Seto, Karen C., Burak Güneralp, and Lucy R. Hutyra. 2012. “Global Forecasts of Urban Expansion to 2030 and Direct Impacts on Biodiversity and Carbon Pools.” 291 Proceedings of the National Academy of Sciences 109 (40): 16083–88. doi:10.1073/pnas.1211658109. Smith, Stanley D., Therese N. Charlet, Stephen F. Zitzer, Scott R. Abella, Cheryl H. Vanier, and Travis E. Huxman. 2014. “Long-Term Response of a Mojave Desert Winter Annual Plant Community to a Whole-Ecosystem Atmospheric CO2 Manipulation (FACE).” Global Change Biology 20 (3): 879–92. doi:10.1111/gcb.12411. Song, Tao, and Yuesi Wang. 2012. “Carbon Dioxide Fluxes from an Urban Area in Beijing.” Atmospheric Research 106 (March): 139–49. doi:10.1016/j.atmosres.2011.12.001. Sponseller, Ryan A., Sharon J. Hall, David P. Huber, Nancy B. Grimm, Jason P. Kaye, Christopher M. Clark, and Scott L. Collins. 2012. “Variation in Monsoon Precipitation Drives Spatial and Temporal Patterns of Larrea Tridentata Growth in the Sonoran Desert.” Functional Ecology 26 (3): 750–58. doi:10.1111/j.1365- 2435.2012.01979.x. Templer, Pamela H. 2013. “Biogeochemistry: Limits on Carbon Uptake by Plants.” Nature Climate Change 3 (3): 184–85. doi:10.1038/nclimate1841. United Nations. 2012. “World Urbanization Prospects, The 2011 Revision”. United Nations publication. Vourlitis, George L. 2012. “Aboveground Net Primary Production Response of Semi- Arid Shrublands to Chronic Experimental Dry-Season N Input.” Ecosphere 3 (3): art22. doi:10.1890/ES11-00339.1. Wang, Ge, and Martin Ostoja-Starzewski. 2004. “Influence of Topography on the Phoenix CO2 Dome: A Computational Study.” Atmospheric Science Letters 5 (5): 103–7. doi:10.1002/asl.67. Watson, John G., Judith C. Chow, Fred W. Lurmann, and Stefan P. Musarra. 1994. “Ammonium Nitrate, Nitric Acid, and Ammonia Equilibrium in Wintertime Phoenix, Arizona.” Air & Waste 44 (4): 405–12. doi:10.1080/1073161X.1994.10467262. Wentz, Elizabeth A., Patricia Gober, Robert C. Balling, and Thomas A. Day. 2002. “Spatial Patterns and Determinants of Winter Atmospheric Carbon Dioxide Concentrations in an Urban Environment.” Annals of the Association of American Geographers 92 (1): 15–28. doi:10.1111/1467-8306.00277. Wesely, M. L, and B. B Hicks. 2000. “A Review of the Current Status of Knowledge on 292 Dry Deposition.” Atmospheric Environment 34 (12–14): 2261–82. doi:10.1016/S1352-2310(99)00467-7. Zhang, L., J. R. Brook, and R. Vet. 2003. “A Revised Parameterization for Gaseous Dry Deposition in Air-Quality Models.” Atmospheric Chemistry and Physics 3 (6): 2067–82 . CHAPTER 5 Aber, John D., W. McDowell, K. Nadelhoffer, A. Magill, G. Berntson, M. Kamakea, S. McNulty, W. Currie, L. Rustad, and Ivan Fernandez. 1998. “Nitrogen Saturation in Temperate Forests: Hypotheses Revisited.” Bioscience 48 (11): 921–34. Aber, John D., Scott V. Ollinger, Charles T. Driscoll, Gene E. Likens, Richard T. Holmes, Rita J. Freuder, and Christine L. Goodale. 2002. “Inorganic Nitrogen Losses from a Forested Ecosystem in Responseto Physical, Chemical, Biotic,and Climatic Perturbations.” Ecosystems 5 (7): 0648–58. doi:10.1007/s10021-002- 0203-8. Ainsworth, Elizabeth A., Craig R. Yendrek, Stephen Sitch, William J. Collins, and Lisa D. Emberson. 2012. “The Effects of Tropospheric Ozone on Net Primary Productivity and Implications for Climate Change*.” Annual Review of Plant Biology 63 (1): 637–61. doi:10.1146/annurev-arplant-042110-103829. Akimoto, Hajime. 2003. “Global Air Quality and Pollution.” Science 302 (5651): 1716– 19. doi:10.1126/science.1092666. Angert, A. L., T. E. Huxman, G. A. Barron-Gafford, K. L. Gerst, and D. L. Venable. 2007. “Linking Growth Strategies to Long-Term Population Dynamics in a Guild of Desert Annuals.” Journal of Ecology 95 (2): 321–31. doi:10.1111/j.1365- 2745.2006.01203.x. Angert, A. L., Travis E. Huxman, Peter Chesson, and D. Lawrence Venable. 2009. “Functional Tradeoffs Determine Species Coexistence via the Storage Effect.” Proceedings of the National Academy of Sciences 106 (28): 11641–45. doi:10.1073/pnas.0904512106. Báez, S., J. Fargione, D. I. Moore, S. L. Collins, and J. R. Gosz. 2007. “Atmospheric Nitrogen Deposition in the Northern Chihuahuan Desert: Temporal Trends and Potential Consequences.” Journal of Arid Environments 68 (4): 640–51. doi:10.1016/j.jaridenv.2006.06.011. 293 Bassin, Seraina, Roland A. Werner, Karin Sörgel, Matthias Volk, Nina Buchmann, and Jürg Fuhrer. 2009. “Effects of Combined Ozone and Nitrogen Deposition on the in Situ Properties of Eleven Key Plant Species of a Subalpine Pasture.” Oecologia 158 (4): 747–56. doi:10.1007/s00442-008-1191-y. Bergeron, Onil, and Ian B. Strachan. 2011. “CO2 Sources and Sinks in Urban and Suburban Areas of a Northern Mid-Latitude City.” Atmospheric Environment 45 (8): 1564–73. doi:10.1016/j.atmosenv.2010.12.043. Bobbink, Roland, K. Hicks, J. Galloway, T. Spranger, R. Alkemade, M. Ashmore, M. Bustamante, S. Cinderby, E. Davidson, and F. Dentener. 2010. “Global Assessment of Nitrogen Deposition Effects on Terrestrial Plant Diversity: A Synthesis.” Ecological Applications 20 (1): 30–59. Brooks, Matthew L. 2003. “Effects of Increased Soil Nitrogen on the Dominance of Alien Annual Plants in the Mojave Desert.” Journal of Applied Ecology 40 (2): 344–53. doi:10.1046/j.1365-2664.2003.00789.x. Bytnerowicz, A., D. M. Olszyk, C. A. Fox, P. J. Dawson, G. Kats, C. L. Morrison, and J. Wolf. 1988. “Responses of Desert Annual Plants to Ozone and Water Stress in an in Situ Experiment.” JAPCA 38 (9): 1145–51. doi:10.1080/08940630.1988.10466463. Cardoso‐Vilhena, João, Luis Balaguer, Derek Eamus, John Ollerenshaw, and Jeremy Barnes. 2004. “Mechanisms Underlying the Amelioration of O3‐induced Damage by Elevated Atmospheric Concentrations of CO2.” Journal of Experimental Botany 55 (397): 771–81. doi:10.1093/jxb/erh080. Cardoso‐Vilhena, João, and Jeremy Barnes. 2001. “Does Nitrogen Supply Affect the Response of Wheat (Triticum Aestivum Cv. Hanno) to the Combination of Elevated CO2 and O3?” Journal of Experimental Botany 52 (362): 1901–11. doi:10.1093/jexbot/52.362.1901. Chapin, F. Stuart, III. 1980. “The Mineral Nutrition of Wild Plants.” Annual Review of Ecology and Systematics 11 (January): 233–60. ———. 1991. “Integrated Responses of Plants to Stress.” BioScience 41 (1): 29–36. doi:10.2307/1311538. Clark, Christopher M., and David Tilman. 2008. “Loss of Plant Species after Chronic Low-Level Nitrogen Deposition to Prairie Grasslands.” Nature 451 (7179): 712– 15. doi:10.1038/nature06503. Crous, Kristine Y., Peter B. Reich, Mark D. Hunter, and David S. Ellsworth. 2010. 294 “Maintenance of Leaf N Controls the Photosynthetic CO2 Response of Grassland Species Exposed to 9 Years of Free-Air CO2 Enrichment.” Global Change Biology 16 (7): 2076–88. doi:10.1111/j.1365-2486.2009.02058.x. Dentener, F., D. Stevenson, K. Ellingsen, T. van Noije, M. Schultz, M. Amann, C. Atherton, et al. 2006. “The Global Atmospheric Environment for the Next Generation.” Environmental Science & Technology 40 (11): 3586–94. doi:10.1021/es0523845. Dijkstra, Feike A., Dana Blumenthal, Jack A. Morgan, Daniel R. LeCain, and Ronald F. Follett. 2010. “Elevated CO2 Effects on Semi-Arid Grassland Plants in Relation to Water Availability and Competition.” Functional Ecology 24 (5): 1152–61. doi:10.1111/j.1365-2435.2010.01717.x. Dukes, Jeffrey S, Nona R Chiariello, Elsa E Cleland, Lisa A Moore, M. Rebecca Shaw, Susan Thayer, Todd Tobeck, Harold A Mooney, and Christopher B Field. 2005. “Responses of Grassland Production to Single and Multiple Global Environmental Changes.” PLoS Biol 3 (10): e319. doi:10.1371/journal.pbio.0030319. Elser, James J., Matthew E.S. Bracken, Elsa E. Cleland, Daniel S. Gruner, W. Stanley Harpole, Helmut Hillebrand, Jacqueline T. Ngai, Eric W. Seabloom, Jonathan B. Shurin, and Jennifer E. Smith. 2007. “Global Analysis of Nitrogen and Phosphorus Limitation of Primary Producers in Freshwater, Marine and Terrestrial Ecosystems.” Ecology Letters 10 (12): 1135–42. doi:10.1111/j.1461- 0248.2007.01113.x. Fangmeier, A, L De Temmerman, C Black, K Persson, and V Vorne. 2002. “Effects of Elevated CO2 And/or Ozone on Nutrient Concentrations and Nutrient Uptake of Potatoes.” European Journal of Agronomy 17 (4): 353–68. doi:10.1016/S1161- 0301(02)00071-0. Felzer, Benjamin S., Timothy W. Cronin, Jerry M. Melillo, David W. Kicklighter, and C. Adam Schlosser. 2009. “Importance of Carbon-Nitrogen Interactions and Ozone on Ecosystem Hydrology during the 21st Century.” Journal of Geophysical Research: Biogeosciences 114 (G1): G01020. doi:10.1029/2008JG000826. Fenn, Mark E., E. B. Allen, S. B. Weiss, S. Jovan, L. H. Geiser, G. S. Tonnesen, R. F. Johnson, et al. 2010. “Nitrogen Critical Loads and Management Alternatives for N-Impacted Ecosystems in California.” Journal of Environmental Management 91 (12): 2404–23. doi:10.1016/j.jenvman.2010.07.034. George, K., L. H. Ziska, J. A. Bunce, and B. Quebedeaux. 2007. “Elevated Atmospheric CO2 Concentration and Temperature across an Urban–rural Transect.” Atmospheric Environment 41 (35): 7654–65. 295 doi:10.1016/j.atmosenv.2007.08.018. Gregg, Jillian W., Clive G. Jones, and Todd E. Dawson. 2003. “Urbanization Effects on Tree Growth in the Vicinity of New York City.” Nature 424 (6945): 183–87. doi:10.1038/nature01728. Gremer, Jennifer R., Sarah Kimball, A. L. Angert, D. Lawrence Venable, and Travis E. Huxman. 2012. “Variation in Photosynthetic Response to Temperature in a Guild of Winter Annual Plants.” Ecology 93 (12): 2693–2704. doi:10.1890/12-0006.1. Grimm, Nancy B., Stanley H. Faeth, Nancy E. Golubiewski, Charles L. Redman, Jianguo Wu, Xuemei Bai, and John M. Briggs. 2008. “Global Change and the Ecology of Cities.” Science 319 (5864): 756–60. doi:10.1126/science.1150195. Groffman, Peter M., Jill S. Baron, Tamara Blett, Arthur J. Gold, Iris Goodman, Lance H. Gunderson, Barbara M. Levinson, et al. 2006. “Ecological Thresholds: The Key to Successful Environmental Management or an Important Concept with No Practical Application?” Ecosystems 9 (1): 1–13. doi:10.1007/s10021-003-0142-z. Hall, Sharon J., Ryan A. Sponseller, Nancy B. Grimm, David Huber, Jason P. Kaye, Christopher Clark, and Scott L. Collins. 2011. “Ecosystem Response to Nutrient Enrichment across an Urban Airshed in the Sonoran Desert.” Ecological Applications 21 (3): 640–60. doi:10.1890/10-0758.1. Heagle, A. S., J. E. Miller, F. L. Booker, and W. A. Pursley. 1999. “Ozone Stress, Carbon Dioxide Enrichment, and Nitrogen Fertility Interactions in Cotton.” Crop Science 39 (3): 731. doi:10.2135/cropsci1999.0011183X003900030021x. Henriques, Fernando S. 2009. “Leaf Chlorophyll Fluorescence: Background and Fundamentals for Plant Biologists.” The Botanical Review 75 (3): 249–70. doi:10.1007/s12229-009-9035-y. Henry, Hugh A. L., Nona R. Chiariello, Peter M. Vitousek, Harold A. Mooney, and Christopher B. Field. 2006. “Interactive Effects of Fire, Elevated Carbon Dioxide, Nitrogen Deposition, and Precipitation on a California Annual Grassland.” Ecosystems 9 (7): 1066–75. doi:10.1007/s10021-005-0077-7. Horowitz, Larry W. 2006. “Past, Present, and Future Concentrations of Tropospheric Ozone and Aerosols: Methodology, Ozone Evaluation, and Sensitivity to Aerosol Wet Removal.” Journal of Geophysical Research: Atmospheres 111 (D22): n/a– n/a. doi:10.1029/2005JD006937. Housman, David C., Elke Naumburg, Travis E. Huxman, Therese N. Charlet, Robert S. Nowak, and Stanley D. Smith. 2006. “Increases in Desert Shrub Productivity 296 under Elevated Carbon Dioxide Vary with Water Availability.” Ecosystems 9 (3): 374–85. doi:10.1007/s10021-005-0124-4. Huxman, Travis E., Greg Barron-Gafford, Katharine L. Gerst, Amy L. Angert, Anna P. Tyler, and D. Lawrence Venable. 2008. “Photosynthetic Resource Use Efficiency and Demographic Variability in Desert Winter Annual Plants.” Ecology 89 (6): 1554–63. doi:10.1890/06-2080.1. Idso, Craig D, Sherwood B Idso, and Robert C Balling Jr. 2001. “An Intensive Two- Week Study of an Urban CO2 Dome in Phoenix, Arizona, USA.” Atmospheric Environment 35 (6): 995–1000. doi:10.1016/S1352-2310(00)00412-X. IPCC. 2014. “Intergovernmental Panel on Climate Change: Fifth Assessment Report.” http://www.ipcc.ch/. Karnosky, David F., John M. Skelly, Kevin E. Percy, and Art H. Chappelka. 2007. “Perspectives Regarding 50 Years of Research on Effects of Tropospheric Ozone Air Pollution on US Forests.” Environmental Pollution 147 (3): 489–506. doi:10.1016/j.envpol.2006.08.043. Kimball, Sarah, Jennifer R. Gremer, Amy L. Angert, Travis E. Huxman, and D. Lawrence Venable. 2012. “Fitness and Physiology in a Variable Environment.” Oecologia 169 (2): 319–29. doi:10.1007/s00442-011-2199-2. Krupa, S. V. 2003. “Effects of Atmospheric Ammonia (NH3) on Terrestrial Vegetation: A Review.” Environmental Pollution 124 (2): 179–221. doi:10.1016/S0269- 7491(02)00434-7. Langley, J. Adam, and J. Patrick Megonigal. 2010. “Ecosystem Response to Elevated CO2 Levels Limited by Nitrogen-Induced Plant Species Shift.” Nature 466 (7302): 96–99. doi:10.1038/nature09176. LeBauer, David S., and Kathleen K. Treseder. 2008. “NITROGEN LIMITATION OF NET PRIMARY PRODUCTIVITY IN TERRESTRIAL ECOSYSTEMS IS GLOBALLY DISTRIBUTED.” Ecology 89 (2): 371–79. doi:10.1890/06-2057.1. Lindroth, Richard L. 2010. “Impacts of Elevated Atmospheric CO2 and O3 on Forests: Phytochemistry, Trophic Interactions, and Ecosystem Dynamics.” Journal of Chemical Ecology 36 (1): 2–21. doi:10.1007/s10886-009-9731-4. Lohse, Kathleen A., Diane Hope, Ryan Sponseller, Jonathan O. Allen, and Nancy B. Grimm. 2008. “Atmospheric Deposition of Carbon and Nutrients across an Arid Metropolitan Area.” Science of The Total Environment 402 (1): 95–105. doi:10.1016/j.scitotenv.2008.04.044. 297 Lombardozzi, D., J. P. Sparks, and G. Bonan. 2013. “Integrating O3 Influences on Terrestrial Processes: Photosynthetic and Stomatal Response Data Available for Regional and Global Modeling.” Biogeosciences Discussions 10 (4): 6973–7012. doi:10.5194/bgd-10-6973-2013. Long, Stephen P, Elizabeth A Ainsworth, Alistair Rogers, and Donald R Ort. 2004. “Rising Atmospheric Carbon Dioxide: Plants FACE the Future.” Annual Review of Plant Biology 55: 591–628. doi:10.1146/annurev.arplant.55.031903.141610. Lovett, Gary M., Timothy H. Tear, David C. Evers, Stuart E.G. Findlay, B. Jack Cosby, Judy K. Dunscomb, Charles T. Driscoll, and Kathleen C. Weathers. 2009. “Effects of Air Pollution on Ecosystems and Biological Diversity in the Eastern United States.” Annals of the New York Academy of Sciences 1162 (1): 99–135. doi:10.1111/j.1749-6632.2009.04153.x. Luo, Yiqi, and James F. Reynolds. 1999. “VALIDITY OF EXTRAPOLATING FIELD CO2 EXPERIMENTS TO PREDICT CARBON SEQUESTRATION IN NATURAL ECOSYSTEMS.” Ecology 80 (5): 1568–83. doi:10.1890/0012- 9658(1999)080[1568:VOEFCE]2.0.CO;2. Maxwell, Kate, and Giles N. Johnson. 2000. “Chlorophyll Fluorescence—a Practical Guide.” Journal of Experimental Botany 51 (345): 659–68. doi:10.1093/jexbot/51.345.659. Melillo, Jm, Ad Mcguire, Dw Kicklighter, B. Moore, Cj Vorosmarty, and Al Schloss. 1993. “Global Climate-Change and Terrestrial Net Primary Production.” Nature 363 (6426): 234–40. doi:10.1038/363234a0. Monks, P. S., C. Granier, S. Fuzzi, A. Stohl, M. L. Williams, H. Akimoto, M. Amann, et al. 2009. “Atmospheric Composition Change – Global and Regional Air Quality.” Atmospheric Environment, ACCENT Synthesis, 43 (33): 5268–5350. doi:10.1016/j.atmosenv.2009.08.021. Morgan, J. A., D. E. Pataki, C. Körner, H. Clark, S. J. Del Grosso, J. M. Grünzweig, A. K. Knapp, et al. 2004. “Water Relations in Grassland and Desert Ecosystems Exposed to Elevated Atmospheric CO2.” Oecologia 140 (1): 11–25. doi:10.1007/s00442-004-1550-2. Ollinger, Scott V., John D. Aber, Peter B. Reich, and Rita J. Freuder. 2002. “Interactive Effects of Nitrogen Deposition, Tropospheric Ozone, Elevated CO2 and Land Use History on the Carbon Dynamics of Northern Hardwood Forests.” Global Change Biology 8 (6): 545–62. doi:10.1046/j.1365-2486.2002.00482.x. Padgett, Pamela E., Andrzej Bytnerowicz, Philip J. Dawson, George H. Riechers, and 298 Dennis R. Fitz. 2004. “Design, Evaluation and Application of a Continuously Stirred Tank Reactor System for Use in Nitric Acid Air Pollutant Studies.” Water, Air, and Soil Pollution 151 (1-4): 35–51. doi:10.1023/B:WATE.0000009890.74470.fa. Pardo, Linda H., Mark E. Fenn, Christine L. Goodale, Linda H. Geiser, Charles T. Driscoll, Edith B. Allen, Jill S. Baron, et al. 2011. “Effects of Nitrogen Deposition and Empirical Nitrogen Critical Loads for Ecoregions of the United States.” Ecological Applications 21 (8): 3049–82. doi:10.1890/10-2341.1. Payne, Richard J., Nancy B. Dise, Carly J. Stevens, David J. Gowing, Cecilia Duprè, Edu Dorland, Cassandre Gaudnik, et al. 2013. “Impact of Nitrogen Deposition at the Species Level.” Proceedings of the National Academy of Sciences 110 (3): 984– 87. doi:10.1073/pnas.1214299109. Potvin, Catherine, and Liette Vasseur. 1997. “LONG-TERM CO2 ENRICHMENT OF A PASTURE COMMUNITY: SPECIES RICHNESS, DOMINANCE, AND SUCCESSION.” Ecology 78 (3): 666–77. doi:10.1890/0012- 9658(1997)078[0666:LTCEOA]2.0.CO;2. Reich, Peter B. 2009. “Elevated CO2 Reduces Losses of Plant Diversity Caused by Nitrogen Deposition.” Science 326 (5958): 1399–1402. doi:10.1126/science.1178820. Reich, Peter B., and Robert G. Amundson. 1985. “Ambient Levels of Ozone Reduce Net Photosynthesis in Tree and Crop Species.” Science 230 (4725): 566–70. doi:10.1126/science.230.4725.566. Reich, Peter B., and Sarah E. Hobbie. 2013. “Decade-Long Soil Nitrogen Constraint on the CO2 Fertilization of Plant Biomass.” Nature Climate Change 3 (3): 278–82. doi:10.1038/nclimate1694. Reich, Peter B., Sarah E. Hobbie, Tali Lee, David S. Ellsworth, Jason B. West, David Tilman, Johannes M. H. Knops, Shahid Naeem, and Jared Trost. 2006. “Nitrogen Limitation Constrains Sustainability of Ecosystem Response to CO2.” Nature 440 (7086): 922–25. doi:10.1038/nature04486. Reid, Chantal D., and Edwin L. Fiscus. 2008. “Ozone and Density Affect the Response of Biomass and Seed Yield to Elevated CO2 in Rice.” Global Change Biology 14 (1): 60–76. doi:10.1111/j.1365-2486.2007.01472.x. Rustad, Lindsey E. 2008. “The Response of Terrestrial Ecosystems to Global Climate Change: Towards an Integrated Approach.” Science of The Total Environment, BIOGEOCHEMISTRY OF FORESTED ECOSYSTEM - Selected papers from 299 BIOGEOMON, the 5th International Symposium on Ecosystem Behaviour, held at the University of California, Santa Cruz, on June 25–30, 2006, 404 (2–3): 222– 35. doi:10.1016/j.scitotenv.2008.04.050. Sala, O. E., F. Stuart Chapin, Iii, Juan J. Armesto, Eric Berlow, Janine Bloomfield, Rodolfo Dirzo, et al. 2000. “Global Biodiversity Scenarios for the Year 2100.” Science 287 (5459): 1770–74. doi:10.1126/science.287.5459.1770. Sala, O. E., and W. K. Lauenroth. 1982. “Small Rainfall Events: An Ecological Role in Semiarid Regions.” Oecologia 53 (3): 301–4. doi:10.1007/BF00389004. Seto, Karen C., Burak Güneralp, and Lucy R. Hutyra. 2012. “Global Forecasts of Urban Expansion to 2030 and Direct Impacts on Biodiversity and Carbon Pools.” Proceedings of the National Academy of Sciences 109 (40): 16083–88. doi:10.1073/pnas.1211658109. Shaw, M. Rebecca, Erika S. Zavaleta, Nona R. Chiariello, Elsa E. Cleland, Harold A. Mooney, and Christopher B. Field. 2002. “Grassland Responses to Global Environmental Changes Suppressed by Elevated CO2.” Science 298 (5600): 1987–90. doi:10.1126/science.1075312. Shen, Weijun, Jianguo Wu, Nancy B. Grimm, and Diane Hope. 2008. “Effects of Urbanization-Induced Environmental Changes on Ecosystem Functioning in the Phoenix Metropolitan Region, USA.” Ecosystems 11 (1): 138–55. doi:10.1007/s10021-007-9085-0. Smith, Melinda D., Alan K. Knapp, and Scott L. Collins. 2009. “A Framework for Assessing Ecosystem Dynamics in Response to Chronic Resource Alterations Induced by Global Change.” Ecology 90 (12): 3279–89. doi:10.1890/08-1815.1. Smith, Stanley D., Therese N. Charlet, Stephen F. Zitzer, Scott R. Abella, Cheryl H. Vanier, and Travis E. Huxman. 2014. “Long-Term Response of a Mojave Desert Winter Annual Plant Community to a Whole-Ecosystem Atmospheric CO2 Manipulation (FACE).” Global Change Biology 20 (3): 879–92. doi:10.1111/gcb.12411. Smith, Stanley D., Travis E. Huxman, Stephen F. Zitzer, Therese N. Charlet, David C. Housman, James S. Coleman, Lynn K. Fenstermaker, Jeffrey R. Seemann, and Robert S. Nowak. 2000. “Elevated CO2 Increases Productivity and Invasive Species Success in an Arid Ecosystem.” Nature 408 (6808): 79–82. doi:10.1038/35040544. Suding, Katharine N., Sandra Lavorel, F. S. Chapin, Johannes H. C. Cornelissen, Sandra Díaz, Eric Garnier, Deborah Goldberg, David U. Hooper, Stephen T. Jackson, and 300 Marie-Laure Navas. 2008. “Scaling Environmental Change through the Community-Level: A Trait-Based Response-and-Effect Framework for Plants.” Global Change Biology 14 (5): 1125–40. doi:10.1111/j.1365-2486.2008.01557.x. Templer, Pamela H. 2013. “Biogeochemistry: Limits on Carbon Uptake by Plants.” Nature Climate Change 3 (3): 184–85. doi:10.1038/nclimate1841. United Nations. 2012. “World Urbanization Prospects, The 2011 Revision”. United Nations publication. Weiss, Stuart B. 1999. “Cars, Cows, and Checkerspot Butterflies: Nitrogen Deposition and Management of Nutrient-Poor Grasslands for a Threatened Species.” Conservation Biology 13 (6): 1476–86. doi:10.1046/j.1523-1739.1999.98468.x. Wentz, Elizabeth A., Patricia Gober, Robert C. Balling, and Thomas A. Day. 2002. “Spatial Patterns and Determinants of Winter Atmospheric Carbon Dioxide Concentrations in an Urban Environment.” Annals of the Association of American Geographers 92 (1): 15–28. doi:10.1111/1467-8306.00277. Zavaleta, Erika S., M. Rebecca Shaw, Nona R. Chiariello, Brian D. Thomas, Elsa E. Cleland, Christopher B. Field, and Harold A. Mooney. 2003. “GRASSLAND RESPONSES TO THREE YEARS OF ELEVATED TEMPERATURE, CO2, PRECIPITATION, AND N DEPOSITION.” Ecological Monographs 73 (4): 585–604. doi:10.1890/02-4053. Chapter 6 Clair, Thomas A., Douglas Burns, Irma Rosas Pérez, Jules Blais, and Kevin Percy. 2011. “Ecosystems.” In Technical Challenges of Multipollutant Air Quality Management, edited by George M. Hidy, Jeffrey R. Brook, Kenneth L. Demerjian, Luisa T. Molina, William T. Pennell, and Richard D. Scheffe, 139– 229. Springer Netherlands. http://link.springer.com/chapter/10.1007/978-94-007- 0304-9_6. Cook, Elizabeth M., Sharon J. Hall, and Kelli L. Larson. 2012. “Residential Landscapes as Social-Ecological Systems: A Synthesis of Multi-Scalar Interactions between People and Their Home Environment.” Urban Ecosystems 15 (1): 19–52. doi:10.1007/s11252-011-0197-0. Fenn, Mark E., E. B. Allen, S. B. Weiss, S. Jovan, L. H. Geiser, G. S. Tonnesen, R. F. Johnson, et al. 2010. “Nitrogen Critical Loads and Management Alternatives for 301 N-Impacted Ecosystems in California.” Journal of Environmental Management 91 (12): 2404–23. doi:10.1016/j.jenvman.2010.07.034. Gregg, Jillian W., Clive G. Jones, and Todd E. Dawson. 2003. “Urbanization Effects on Tree Growth in the Vicinity of New York City.” Nature 424 (6945): 183–87. doi:10.1038/nature01728. ———. 2006. “PHYSIOLOGICAL AND DEVELOPMENTAL EFFECTS OF O3 ON COTTONWOOD GROWTH IN URBAN AND RURAL SITES.” Ecological Applications 16 (6): 2368–81. doi:10.1890/1051- 0761(2006)016[2368:PADEOO]2.0.CO;2. IPCC. 2014. “Intergovernmental Panel on Climate Change: Fifth Assessment Report.” http://www.ipcc.ch/. Larson, Kelli L., Elizabeth Cook, Colleen Strawhacker, and Sharon J. Hall. 2010. “The Influence of Diverse Values, Ecological Structure, and Geographic Context on Residents’ Multifaceted Landscaping Decisions.” Human Ecology 38 (6): 747–61. doi:10.1007/s10745-010-9359-6. Lovett, Gary M., Timothy H. Tear, David C. Evers, Stuart E.G. Findlay, B. Jack Cosby, Judy K. Dunscomb, Charles T. Driscoll, and Kathleen C. Weathers. 2009. “Effects of Air Pollution on Ecosystems and Biological Diversity in the Eastern United States.” Annals of the New York Academy of Sciences 1162 (1): 99–135. doi:10.1111/j.1749-6632.2009.04153.x. Pardo, Linda H., Mark E. Fenn, Christine L. Goodale, Linda H. Geiser, Charles T. Driscoll, Edith B. Allen, Jill S. Baron, et al. 2011. “Effects of Nitrogen Deposition and Empirical Nitrogen Critical Loads for Ecoregions of the United States.” Ecological Applications 21 (8): 3049–82. doi:10.1890/10-2341.1. Seto, Karen C., Burak Güneralp, and Lucy R. Hutyra. 2012. “Global Forecasts of Urban Expansion to 2030 and Direct Impacts on Biodiversity and Carbon Pools.” Proceedings of the National Academy of Sciences 109 (40): 16083–88. doi:10.1073/pnas.1211658109. United Nations. 2012. “World Urbanization Prospects, The 2011 Revision”. United Nations publication. 302