Why “Feed the Lawn”? Exploring the Influences on Residential Turf Grass Fertilization in the Minneapolis−Saint Paul Metropolitan Area

Abstract

We contribute to scholarly understanding of lawn fertilization behavior by formulating and testing models of fertilization and fertilization frequency that incorporate a wide range of human and structural influences. Specifically, we explore the impact of (a) structural/environmental aspects, (b) sociodemographics, and (c) attitudes and beliefs (informal norms, knowledge, environmental concerns, yard motivations, perceived ability regarding fertilization practices). We use survey data from a sample of 942 residents within the Minneapolis/Saint Paul, Minnesota metropolitan area. Our results indicate that fertilization is not driven by one overriding factor but by a combination of factors. Specifically, we show significant effects from a range of structural aspects (lot size, location, use of a professional lawn service), sociodemographics (home value), norms (perceived neighbor effects), knowledge (perceived positive/negative aspects of fertilization), and perceived ability. We also explore the urban/suburban divide and show higher fertilization stemming from suburban households.

Keywords

lawn care fertilization yard management attitudes norms behavior urban and suburban

The residential yard provides a wide range of effects on the homeowner and the surrounding ecosystem. It is an outlet for social activity, recreation, family connections, and neighborhood pride (e.g., Jenkins, 1994; Larson, Sagrande, Harlan, & Yabiku, 2009; Nielson & Smith, 2005). At the same time, the residential yard provides an economic feature to homeowners and the surrounding neighborhood (e.g., Nassauer, Wang, & Dayrell, 2009). Equally important, the residential yard provides a number of positive and negative environmental effects on the surrounding ecosystem and beyond, providing carbon sequestration, water filtration, and urban heat island mitigation (Bandaranayake, Qian, Parton, Ojima, & Follett, 2003; Beard & Green, 1994; Jenerette et al., 2007; Milesi et al., 2005), while potentially contributing to nutrient runoff and eutrophication of local streams, lakes, as well as coastal zones (e.g., Carpenter et al., 1998; Robbins & Sharp, 2003a).

The negative effects and externalities of the residential landscape have become a significant concern among urban practitioners and scientists alike, and have led to a considerable focus on individual yard care behavior. One specific behavior that has generated attention is the application of fertilizer to residential lawns (e.g., Blaine, Clayton, Robbins, & Grewal, 2012; Carrico, Fraser, & Bazuin, 2013; Jenkins, 1994; Law, Band, & Grove, 2004; Osmond & Hardy, 2004; Robbins & Sharp, 2003a). This attention to fertilizer and the residential yard is due to (a) the growth in urban development and hence more lawn space overall (e.g., Lubowski, Vesterby, Bucholtz, Baez, & Roberts, 2006; Robbins, Polderman, & Birkenholtz, 2001), (b) the substantial increase in fertilization amount and frequency (e.g., Butterfield, 2004; Fissore et al., 2011; Jenkins, 1994), and (c) the negative impact of fertilization on local ecosystems (e.g., Carpenter et al., 1998; Greenlee, Ellis, & Berg, 2004; Robbins & Sharp, 2003a).

The total area devoted to turf grass has increased substantially. Lubowski and colleagues (2006) note that urban land has nearly quadrupled from 15 million acres in 1945 to 60 million acres in 2002. As a consequence of this urban growth, agricultural and open land is replaced with lawns (Robbins et al., 2001). When considering the total size of turf grass cover, Robbins and Bikenholz (2003) estimate that nearly a quarter of urban cover is dedicated to lawns, which amount to 164,000 square kilometers, three times the size of any irrigated crop (Milesi & colleagues, 2005). In comparison with total agriculture, lawns do make a smaller proportion of total land but at the same time, lawns, and the downstream ecosystems they drain into, are very near where people live. Hence, lawns affect very high value ecosystems.

Second, fertilization of residential lawns is an individual behavior that is widespread among the population, substantial, and the amount of fertilizer used and the frequency of fertilization are increasing. Jenkins (1994) notes that fertilization application recommendations increased from one pound of nitrogen per 1,000 square feet in 1940 to eight pounds per 1,000 square feet in the 1970s (Current recommended fertilizations levels are lower but can still reach up to 6 lbs/1,000 sq. ft. for certain lawn characteristics; Rosen, Horgan, & Mugaas, 2006). By 1984, total applied chemical fertilizer was at nearly a million tons annually (Jenkins, 1994). Butterfield (2004) notes that out of 90 million households with a yard or garden, nearly 60% will apply fertilizer to their yard or garden. When considering the share of total nitrogen flowing through the household, fertilization comes in second (behind nitrogen in food consumed; Fissore et al., 2011). For these households, N (nitrogen) fertilization was highly skewed: 28% did not fertilize in the year of the survey, 40% fertilized once or twice, and 32% fertilized three or more times. For all of the 29% of households that fertilized with N and irrigated, total nitrogen inputs (fertilizer + other sources) exceeded ecosystem demand, indicating a high potential for export of nitrogen in runoff.

Third, fertilizer has been connected to a wide range of negative impacts on the environment (e.g., contaminated streams, rivers, lakes), wildlife, and humans. A wide range of chemicals are used on the residential lawn, and these chemicals have a variety of toxic effects (Greenlee et al., 2004; Robbins & Sharp, 2003a). Beyond toxicity, lawn fertilization is a potential source of nitrogen to downstream ecosystems, with a range of potential negative effects, for example, lake eutrophication (Carpenter et al., 1998).

Essentially, while the average individual may perceive fertilization as a benefit to the residential lawn, fertilization may have a wide range of negative influences on the environment. These negative effects have become amplified due to the increasing rate of fertilization on the residential lawn and also the overall growth of the residential lawn in the United States. These concerns are largely behind scholarship’s attempt to understand the factors driving fertilization practices (e.g., Blaine et al., 2012; Carrico et al., 2013; Osmond & Platt, 2000; Templeton, Yoo, & Zilberman, 1999) and are the driving force behind our analysis as well.

Our analysis draws from and builds on previous analyses in a number of ways. First, we conduct an extensive analysis of a wide range of influences on lawn fertilization, such as structural factors, demographics, and individual beliefs (affective and cognitive components). While recent literature has been more frequent in considering the drivers of lawn fertilization (e.g., Blaine et al., 2012; Carrico et al., 2013), this area is still heavily understudied when compared with other areas such as water use or urban biodiversity. Moreover, current studies of fertilization typically focus on a few drivers where we provided an extensive list of drivers. By providing such an extensive analysis, we are able to consider the strongest influences across a range of potentially relevant factors. Second, while research has consistently explored the influence on fertilization (whether the individual fertilizes or not), we also consider influences on the frequency of fertilization. Knowing what drives fertilization itself is quite important, but it is also vital to understand what drives an individual to fertilize once per year compared with someone who fertilizes six times per year. For example, given the heavily skewed nature of N contribution due to fertilization (Fissore et al., 2011) identifying influences of “high fertilizers” is useful to formulating policies around decreasing overall N contribution. Third, this analysis provides an urban versus suburban comparison and considers how influences change depending on the location. Some research has noted differences in yard management behavior depending on location (Blaine et al., 2012) and given the continued growth of urban and suburban communities, it is important to further understand how different community sizes lead to changes in lawn management behavior. Finally, this adds a new region, the Minneapolis/St. Paul, Minnesota metropolitan area, to the growing literature on fertilization of lawns. Minnesota provides a northern midwest study site to the growing list of locations (e.g., Arizona, Maryland, Ohio, Tennesse, Florida; also see Chowdhury, Larson, Grove, Polsky, & Cook, 2011, for a discussion on homogenization across study sites), and allows for continued exploration of the generalizability of theories and results across the nation.

In the following pages, we take up this continued exploration into fertilization practices. First, we begin with a theoretical discussion regarding the factors that should influence fertilization practices. Next we move to a quantitative analysis of these factors. Finally, we discuss the results of our analysis, as well as the implications of our findings.

Fertilization Influences

Theoretical and empirical explorations into fertilization practices and general landscape management decisions have uncovered a wide range of influences (see Cook, Hall, & Larson, 2012, for a review). Many of these landscape management influences are driven uniquely by a combination of human drivers (e.g., knowledge, sociodemographics, beliefs, preferences, concerns) and also ecological drivers (e.g., location, ground cover, climate, soil characteristics). There is not one overriding influence on lawn and fertilizer management decisions so we need to consider a wide range of influences. At the same time, we should assess these different factors in the same empirical models to compare their influence in relation to each other. To that end, we consider a range of influences, consisting of (a) structural/environmental aspects, (b) sociodemographics, and (c) individual beliefs. In essence, we propose the following model of fertilization practices and elaborate below:

Fertilization Practices = f (structure + sociodemographics + beliefs) .

Structural Aspects

First the role of structural aspects on fertilization practices consists of more defined aspects of the property itself. Examples of these aspects include ground cover, property size, age of property, lawn type, and location. These aspects have been shown to be quite influential on landscape practices (although some results have been mixed). For example, a number of studies have indicated property/house age to be a significant influence on fertilizer application (Law et al., 2004; Zhou, Troy, & Grove, 2008; Zhou, Troy, Grove, & Jenkins, 2009). Law and colleagues (2004) considered fertilizer application rates in two areas within Baltimore and noted that newer properties had higher fertilizer application rates. These results are complimented by Zhao and colleagues (2008) who also noted higher fertilization rates for newer properties. Lawn area has also been a structural factor that has received substantial investigation. Zhao and colleagues (2008) found that homeowners with larger lawns applied more total fertilizer but when considering the total size of the lawn area the rate was less. In contrast, Law et al. (2004) did not find a significant effect from lawn size. Other structural factors have also played a role in fertilization practices. Blaine and colleagues (2012) noted that rural homeowners were less likely to fertilize than urban and suburban homeowners. Moreover, professional lawn company use has been shown to lead to higher fertilization application (Fissore et al., 2011; Law et al., 2004; Osmond & Platt, 2000).

Sociodemographics

Sociodemographic traits have also been shown to have a substantial impact on landscape preferences. Of these factors, income has been one of the strongest predictors of increased fertilization use (Blaine et al., 2012; Law et al., 2004; Robbins et al., 2001; Osmond & Hardy, 2004; Osmond & Platt, 2000; Templeton et al., 1999). While these results tend to show a positive relationship (increasing income equals increased fertilizer use), others have shown a nonlinear effect. Law and colleagues (2004) demonstrated that home value (a strong proxy for income) had a relationship with fertilization but that the relationship was more of an inverted U and that the highest fertilization application rate was actually found within the middle income group. Age has also been shown to be a significant predictor. Robbins and colleagues (2001) showed that age was positively correlated with chemical (including fertilizer) use (higher probability of chemical use among older individuals) where Templeton et al. (1999) found a nonlinear effect (probability of chemical usage increased until ages 40 to 45 and then decreased). Gender, education, children, and pets in the household have also been shown to influence fertilization practices (Blaine et al., 2012; Carrico et al., 2013; Robbins et al., 2001; Templeton et al., 1999). For example, Robbins and colleagues (2001) found a positive correlation between education and chemical usage. Blaine and colleagues (2012) found women to be more likely to employ a lawn company to apply chemicals to their lawn, but there was no statistical difference between men and women in applying chemicals themselves. Finally, Carrico and colleagues (2013) found a positive connection between having pets and children and fertilizer usage.

Beliefs and Attitudes

Beyond structural and sociodemographic influences, we also need to consider how differences in individual beliefs and attitudes influence fertilization practices. A wide range of scholarship has considered how psychological and sociological components can contribute to our understanding of environmental problems and yard management (Carrico et al., 2013; Dunlap & Jones, 2002; Larson, Cook, Strawhacker, & Hall, 2010; Larson, Ibes, & White, 2011; Stern, Dietz, Abel, Guagnano, & Kalof, 1999; see also Cook et al., 2012). Management decisions are not driven simply by some environmental feature but also by homeowner beliefs, values, norms, concerns, and preferences regarding his or her yard and how best to create the preferred yard. Following recent literature on the conceptualization and operationalization of beliefs and attitudes (Dunlap & Jones, 2002; Larson, Ibes, & White, 2011), we group these influences as either cognitive components (technical knowledge, norms, ability) or affective components (motivations and concerns).

Cognitive Aspects

Cognitive components reference the more knowledge-based judgments of human understanding (factual or subjective). These components focus on a level of subjective understanding, beliefs, or awareness (Dunlap & Jones, 2002; Larson et al., 2011). For cognitive components, we consider (a) specific technical knowledge or belief around fertilization, (b) beliefs in some normative influence around fertilization and yard management, and (c) belief in the individual’s ability to fertilize. Let us first consider an individual’s technical knowledge or belief. Certainly there is a knowledge gap when it comes to some issues of landscape management (Carrico et al., 2013; Nielson & Smith, 2005), but regardless of the level of knowledge, an individual’s belief regarding fertilization and its impact could have a significant effect on his or her fertilization practices. Beliefs help individuals assess which actions are positive or negative and this assessment directly influences intentions and actions (Ajzen, 1991; Fishbein & Ajzen, 1975). So, for example, if individuals believe it is harmful to the environment, then they will have a more negative assessment of fertilizer and be less likely to use it. On the other hand, if individuals believe fertilizer is beneficial to lawns, then they have a more positive assessment and are more likely to use it.

Empirical results are somewhat mixed regarding the role of technical knowledge/beliefs on landscape management and fertilization practices. Robbins and colleagues (2001) noted that, of individuals who stated their lawn care practices created negative environmental effects, 74% used chemicals compared with only 41% of individuals who said their lawn practices had no effect. A similar positive effect between concerns and chemical usage was found by Blaine and colleagues (2012). Robbins and Sharp (2003b) inquired how beliefs in the safety of fertilization impacted practices. They found that respondents who believed that lawn chemicals were unsafe were less likely to use a lawn care company to apply chemicals to their yard. However, this relationship disappears when asking respondents about using chemicals themselves.

A second cognitive component is the belief in some neighborhood norm or expectation regarding a well-kept yard. While further study of norms along with formal and informal institutions is needed, the current literature has pointed to the power of norms and informal institutions in influencing landscape practices and fertilization (Blaine et al., 2012; Carrico et al., 2013; Nielson & Smith, 2005; Robbins et al., 2001; Robbins & Sharp, 2003b; ). To some individuals, the idea of a well-kept lawn is an internalized moral obligation (Jenkins, 1994; Robbins et al., 2001); to others, the norm is pushed for economic reasons (Feagan & Ripmeester, 1999; Nassauer et al., 2009); and to others, it a source of personal pride (Jenkins, 1994; Nielson & Smith, 2005). In any case, substantial research, considering qualitative and quantitative data, has found a strong normative effect emanating from neighborhoods. For example, recent work by Carrico and colleagues (2013) showed a significant and positive relationship between social pressures and fertilizer use in a sample in Nashville, TN. When analyzing personal interviews with individuals in Oregon, Nielson and Smith (2005) note individual statements such as: “I like to keep my yard looking nice. The rest of the neighborhood would hate you if you didn’t” and “Around here it is a competition! I think it is important for everyone to keep their yards nice.”

Finally, one additional cognitive aspect is the individual’s perceived ability to actually fertilize his or her lawn. Individuals may be motivated by their notion of an ideal lawn and believe fertilization is important and good, but if he or she lacks the ability (or perceived behavioral control; Ajzen, 1991, 2002) to actually fertilize, then that behavior is less likely to occur. In general, we may perceive fertilization as a simple and easy task for many, but there are a wide range of situations that may lead individuals to believe they do not have the ability or control (e.g., time, physical, or even monetary constraints). When assessing individual preferences of their ideal yard, it is not uncommon for the theme of “ease of maintenance” to play a role (Blaine et al., 2012; Larson et al., 2009). Tasks that are perceived as “easy” are much more likely to occur compared with more difficult tasks.

Affective Components

While cognitive aspects of belief are important considerations for fertilization usage, we should also consider affective components. Affective components reference feelings, values, and emotions and don’t have to be tied to actual knowledge regarding a subject (Dunlap & Jones, 2002; Larson et al., 2011). For example, Larson and colleagues (2011) argue that “expressions of concern are distinctive” as emotional responses regarding an issue or action. With this, we consider two affective components: (a) environmental concerns around local waters and (b) lawn care motivations. Environmental concerns imply an emotion judgment or personal worry (Blaine et al., 2012; Dunlap & Jones, 2002; Larson et al., 2011) toward a specific problem. In the case of fertilizing, we would expect individuals with greater environmental concerns for local waters to be less likely to fertilize. The emotional concern should lead individuals to follow practices that are perceived to be less harmful to local waters (less fertilizing). Empirically, however, there is mixed support for this notion. Feagan and Ripmeester (1999) noted that individuals who were more environmentally conscious were less likely to use fertilizers, where Templeton and colleagues (1999) found that membership in an environmental group led to more chemical use.

Another factor to consider is an individual’s motivation for maintaining his or her lawn. Landscape choices are driven by an individual’s preferences and ideal image of his or her yard (Carrico et al., 2013; Larson et al., 2009; Nielson & Smith, 2005; Robbins & Sharp, 2003a). Individuals who imagine their yards as places for recreation may prefer a more open and grass-filled yard, whereas individuals who see their yards as a place of comfort and seclusion may prefer more trees. To achieve that ideal yard image, individuals need to make landscape choices. These choices will include a wide range of actions, and fertilizer usage is one of them. For example, Larson and colleagues (2009) noted the dominant reasons for landscape choices revolved around factors such as appearance, health and safety, recreation, and maintenance. In specifically looking at fertilization usage, Carrico and colleagues (2013) found that individuals’ interests with lawn care decisions (as measured by perceptions of effect of lawn on home value, importance of a green lawn, and impact of lawn on individual reputation) were significant predictors for increased fertilizer use.

In summary, we hypothesize that fertilization and fertilization frequency are driven by a range of influences, consisting of (a) structural/environmental aspects, (b) sociodemographics, and (c) individual beliefs. We now move to test these hypotheses.

Method

To explore the influences on fertilization practices, we turn to a lawn and yard care study, conducted by researchers at the University of Minnesota. The Lawn and Yard Care Choices survey (Nelson et al., 2011) was conducted in April of 2011 and is part of the Twin Cities Household Ecosystem Project (TCHEP). Mail surveys were randomly sent to 2,000 households in the Highland Park neighborhood of St. Paul, Minnesota, and the city of Lino Lakes, Minnesota (4,000 total surveys). Lino Lakes, Minnesota, is part of the Rice Creek Watershed District and Highland Park is part of the Capitol Region Watershed District. Both are part of the Minneapolis/St. Paul metropolitan area, but also provide unique insights and comparisons between a more urban (Highland Park) and suburban (Lino Lakes) setting. The study collected quantitative and qualitative data on social/psychological, socioeconomic, and biophysical variables that potentially influence homeowners’ decision making about yard care as well as household-level nutrient fluxes through a mixed-methods approach. The overall study consisted of a mail survey, small group discussions, and mailed information exchanges with homeowners. The focus of this analysis is the initial contact with respondents (the mail survey). The mail survey consisted of 55 questions and took approximately 20 minutes to complete.

Survey Sample Selection

The survey population frame was single-family, detached, resident-managed households (of these households, 934 are owner-occupied and 8 were renters) within the Highland Park neighborhood in St. Paul, Ramsey County, and Lino Lakes, Anoka County, that did not receive a prior survey (conducted in 2008) by the Twin Cities Household Ecosystem Project (TCHEP). The selection of households to participate in the survey was a three-step process: (a) selection of base data for households, (b) selection of candidate households based on survey population criteria, and (c) random selection of households from candidates to be the survey population.

The MetroGIS Regional Parcel dataset was selected as the geographic base layer for household selection. The parcel dataset is a quarterly compilation of tax parcel polygon layers from the seven Twin Cities area counties; this study used data released in December 2010. The parcel data were subset geographically by the two study areas. The total number of parcels was 6,513 in Highland Park and 7,679 in Lino Lakes.

The parcel data contain several attributes that were used to identify candidate parcels that met the study criteria. Land use and dwelling type attributes identified parcels that were residential, and contained single-family and detached houses. In addition, the homestead attribute was used to find parcels that were owner-occupied. To augment the parcel data, Metropolitan Council 2005 land use data were used to identify and eliminate parcels in areas with land use classifications of multifamily, townhouse, or housing park. Next, the parcels were visually inspected to reveal any parcels that were not identified as detached for tax purposes in the parcels but could be seen visually to be in attached housing complexes (through GIS). Those parcels were eliminated from the selection. Finally, respondents to a previous survey (Fissore et al., 2011; Nelson et al., 2008) were also eliminated from the selection. The remaining parcels (4,671 in Highland Park and 4,861 in Lino Lakes) were candidates to receive the Lawn Care survey.

In each of the study areas, 2,000 of the candidates parcel (4,000 total) were selected randomly to be the Lawn Care survey households. In each study area, the total candidates and the selected households were summarized by census blocks (U.S. Census 2010) to verify that the percentage selected was the same by block. This ensured that the number of addressees selected from each census block was proportional to the density of the block, such that higher density census blocks received more surveys (per unit area) than lower density census blocks.

The Yard Care Choices survey was mailed with an introductory letter explaining the study to selected households in April 2011. It requested that the individual in the household who was most responsible for lawn care complete the survey. About 2 weeks later, we sent a reminder postcard. We received a total of 942 completed surveys, a response rate of 24%. On comparing a few basic sociodemographics of our sample with the general population, we did note some differences in education, income, and age (all somewhat higher in our sample). At the same time, we only had data on the entire population in these areas and not just homeowners, which is likely contributing to some of these differences. Moreover, our sample was 98% White, where it was 90% in Lino Lakes and 79% in Highland Park.

In addition, we did not consider respondents who reported they did not have a lawn (six respondents). Respondents varied considerably in the size and proportion of their yard devoted to the lawn. And while we asked a variety of questions regarding lawn fertilization, we did not ask questions regarding fertilization of other yard vegetation.

Statistical Procedures

A wide range of variables from the Yard Care Choices survey were used to operationalize each of our different factors, along with our dependent variable (fertilization practices). Tables 1 and 2 provide the detailed variable coding, along with summary statistics. Let us begin with the dependent variable. The survey included questions regarding whether households fertilized and at what frequency in 2010. For fertilization, the question asked, “Did you fertilize your lawn in 2010?” The question lead to dichotomous variables (1 = Yes, 0 = No) and is analyzed with logistic regression. For fertilization frequency, the following question was used: “About how many times did your household fertilize your lawn each season in 2010?” Respondents were allowed to insert any number for each of the four seasons. The range was from 1 to 23 times (we only considered respondents who fertilized in 2010). Given the count nature of this variable, we used a Poisson model to analyze this dependent variable. We also considered fertilizer frequency through a second operationalization as well. Using this same variable, we constructed another dependent variable aimed at looking specifically at the high fertilizers. We dichotomized homeowners into “high fertilizers” (four or more fertilizer treatments in 2010) versus all others. While the choice of four fertilizer treatments was somewhat arbitrary, we felt it made sense because it was about one standard deviation above the mean fertilizer frequency.

Table 1.

Variable Descriptions.

Variables	Item description/coding
Dependent variables
Fertilize lawn 2010	Coded 1 if R fertilized lawn in 2010 and 0 otherwise
How many times R fertilized in 2010	1 to 23 scale indicating how many times R fertilized his/her lawn in 2010 (if fertilized in 2010)
High fertilizers	Coded 1 if R fertilized lawn 4+ times in 2010 and 0 otherwise
Structural
Log of lot size (acres)	Log of total lot size
Highland Park	1 if house is in Highland Park and 0 if in Lino Lakes
Use lawn company	1 if R uses a professional lawn care service and 0 otherwise
Sociodemographic
Log of market value	Log of estimated market value
Length of home ownership	Continuous measure of how many years R has owned the house
Pets	Coded 1 if homeowner has pets and 0 otherwise
Children	Coded 1 if children living at address and 0 otherwise
Age	Continuous measure of R’s age
Education	5-point ordinal scale where higher values equal more education
Female	Coded 1 if R is female and 0 if male
Cognitive components
Neighbors think I should not fertilize	5-point ordinal scale where 1 = neighbors think R should fertilize and 5 = neighbors think R should not fertilize
Storm water is treated	Coded 1 if R believes storm water goes to a water treatment facility and 0 otherwise
Fertilization = pollution	5-point ordinal scale where 1 = R believes pollution is unlikely from pollution and 5 = R believes pollution is likely from fertilization
Fertilization ≠ attractive lawn	5-point ordinal scale where 1 = R believe fertilization is likely to create an attractive lawn and 5 = fertilization is unlikely to create an attractive lawn
Ease of fertilization	“For me to fertilize my lawn this year is”: (possible/impossible), (difficult/easy), (feasible/infeasible). Variables are summed, scoring is 0-4, with 4 reflecting ease of fertilization. Alpha = .83.
Affective components
Concerns with local waters	“What concerns, if any, do you have about the water bodies close to your house?” 4 items used: (1) excess nutrients, (2) health of fish and other aquatic life, (3) safety for swimming, (4) safety for fishing. Variables are summed, scoring is 0-4, with 4 reflecting greater concern. Alpha = .85.
Lawn maintenance motivations	“Households maintain lawns for many reasons. How important are the following reasons to you?” 6-point ordinal scale where higher values indicate more importance for each of the reasons.
Children/pets
Relax/enjoy
Attractive prop.
Enjoy
Good neighbor
Prevent erosion

Table 2.

Variable Summary Statistics.

Variables	N	M	SD	Minimum	Maximum
Dependent variables
Fertilize lawn in 2010	935	0.75	0.43	0	1
Fertilize frequency in 2010 (if fertilized)	648	2.69	1.65	1	23
High fertilizers (4+ times in 2010)	884	0.17	0.38	0	1
Structural
Log of lot size (acres)	942	0.24	0.37	0	2.797
Highland Park	942	0.56	0.50	0	1
Use lawn company	923	0.21	0.41	0	1
Sociodemographic
Log of market value	942	12.56	0.37	11.49	14.04
Length of home ownership	828	17.92	12.32	.5	78
Pets	942	0.55	0.50	0	1
Children	896	0.35	0.48	0	1
Age	859	52.78	12.79	24	97
Education	877	3.95	1.04	1	5
Female	872	0.40	0.49	0	1
Cognitive components
Neighbors think I should not fertilize	750	2.76	0.90	1	5
Storm water is treated	911	0.11	0.31	0	1
Fertilization = pollution	880	3.01	1.24	1	5
Fertilization ≠ attractive lawn	894	2.05	1.08	1	5
Ease of fertilization to R	838	3.30	0.99	0	4
Affective components
Concerns with local waters	837	3.69	1.02	1	5
Lawn maintenance motivations
Children/pets	919	3.91	1.90	1	6
Relax/enjoy	928	4.91	1.32	1	6
Attractive property	929	4.85	1.30	1	6
Enjoy	927	3.66	1.63	1	6
Good neighbor	925	4.32	1.26	1	6
Prevent erosion	925	4.15	1.47	1	6

To understand influences on fertilization practices, we considered a wide range of explanatory variables that tapped into the different structural, sociodemographic, normative, belief, motivation, and ability considerations outlined above. We briefly discuss each variable in the text below and provide a more detailed description and coding procedure in Table 1. We considered three aspects of structural variables: (a) homeowner lot size (natural log), (b) use of professional lawn service, and (c) location of either Highland Park or Lino Lakes. Property age was not included in the analysis because it introduced collinearity problems within the models (property age is highly correlated (r = −.84) with location (Highland Park or Lino Lakes) so we chose to maintain the location variable but not the age variable).

As for sociodemographic factors, we used a variety of measures. First, we used the log of the property market value as a measure for the overall household economic status (actual household income was not used because of a substantial number of missing values [nearly 200]. Moreover, household income and the log of market value were highly correlated [r = .51]). Second, we included the length of home ownership. Finally, we included measures for the respondent’s age, education, gender, and the presence of children or pets in the home.

For beliefs, we used a number of different measures. First, we used two measures tapping into knowledge regarding the impact of fertilization on lawn care. One considered if the respondent believed that pollution is likely from fertilization; the other considered if the respondent believed that fertilization led to an attractive lawn. In addition to these questions, we also included a measure of whether the respondent believed his or her storm water was treated by a water treatment facility. For norms, we used a variable that asked if the respondent believed his or her neighbors thought the respondent should or should not fertilize. Finally, we measured the respondent’s perceived ability to fertilize with an additive index of three items—“For me to fertilize my lawn this year is” (5-point scale): (a) possible to impossible, (b) difficult to easy, (c) feasible to infeasible (α = .83). For environmental concerns, we included an additive index of four items—“What concerns, if any, do you have about the water bodies close to your home” (5-point scale): (a) excess nutrients, (b) health of fish and other aquatic life, (c) safety for swimming, (d) safety for fishing (α = .85). Finally, for motivations, we used a battery of questions to measure how important specific reasons for maintaining the lawn were to the respondent. These different reasons are (a) place to play for children and pets, (b) place to relax and be outside, (c) essential to an attractive property, (d) activity I enjoy, (e) being a good neighbor, (f) vegetative cover to prevent soil erosion.

Multivariate Analysis—Understanding Fertilization Practices

Table 3 provides the multivariate models predicting fertilization practices. Model 1 considers if the individual fertilized in 2010 or not, model 2 considers the frequency of fertilization, and model 3 compares “high fertilizers” (4+ fertilizer applications in 2010) with all other respondents. Model 1 of Table 3 indicates that each of the different categories (structural, sociodemographic, and beliefs) is significant, at varying degrees, in influencing fertilization. The model fit is also quite substantial. The adjusted r-square (McFadden) is .46, the percent correctly predicted from the model is 88%, and the proportional reduction in error is 49% (all quite substantial).

Table 3.

Factors Leading to Fertilization in 2010.

	Fertilize (yes/no)		Frequency (only of fertilizers)		High fertilizers (4+ applications)
	Odds ratio	SE	Coefficient	SE	Odds ratio	SE
Structural
Log of lot size (acres)	0.266***	0.122	−.063	0.071	0.455	0.300
Highland Park	0.191****	0.083	−.110**	0.053	0.432**	0.161
Use lawn company	2.311*	1.065	.427****	0.054	7.168****	2.445
Sociodemographic
Log of market value	4.071***	2.184	−.119	0.073	0.629	0.260
Length of home ownership	1.007	0.021	−.001	0.002	1.006	0.015
Pets	1.232	0.374	−.061	0.048	0.616	0.182
Children	1.613	0.577	.055	0.057	1.541	0.569
Age	0.993	0.020	.000	0.004	0.975	0.019
Education	1.014	0.168	.024	0.024	1.323*	0.208
Female	0.667	0.210	−.083*	0.050	0.644	0.215
Cognitive components
Neighbors think I should not fertilize	0.666**	0.120	.043	0.036	1.435*	0.283
Storm water is treated	2.620*	1.491	.053	0.068	1.184	0.593
Fertilization = pollution	0.583****	0.088	−.004	0.024	0.939	0.141
Fertilization ≠ attractive lawn	0.717*	0.123	−.127****	0.030	0.448****	0.109
Ease of fertilization	3.344****	0.672	.166****	0.042	2.826***	0.921
Affective components
Concerns with local waters	0.883	0.140	−.014	0.021	1.026	0.155
Lawn maintenance motivations
Children/pets	1.017	0.096	.009	0.015	0.941	0.096
Relax/enjoy	0.836	0.125	−.013	0.024	1.101	0.171
Attractive prop.	1.209	0.175	.056**	0.028	1.442*	0.295
Enjoy	1.012	0.109	.032*	0.018	1.182	0.140
Good neighbor	1.336*	0.209	−.015	0.026	0.962	0.158
Prevent erosion	1.129	0.126	−.007	0.015	1.001	0.105
Constant	0.000***	0.000	1.631**	0.805	0.253	1.278
N	515		371		496
R²/pseudo R²	.46		.07		.31
Percent correctly predicted	88				85
Proportional reduction in error (%)	49				20

p < .10. **p < .05. ***p < .01. ****p < .001.

First, let us consider the structural factors. All three of these variables are statistically significant and have a moderate substantive effect as well. For the lot size, going from one standard deviation below the mean to one standard deviation above the mean decreases the probability of fertilization from .78 to .62. Location (suburban versus urban) also provides a substantive difference. The probability of fertilizing is .93 in Lino Lakes (suburban) compared with .73 in Highland Park (urban). Finally, the use of a lawn company increases the probability of fertilization from .73 (self-managed) to .86 (professionally managed).

As for the sociodemographic variables, the only variable significant is property value. Its substantive value is also moderate with a .20 increase in the probability of fertilizing when going from one standard deviation below the mean to one standard deviation above the mean.

With beliefs, we show a variety of statistically and substantively significant variables. First let us consider the cognitive items. The norm variable has a large substantive effect. For individuals who responded that their neighbors felt they should fertilize (coded 1), the probability of fertilizing was .85. For those who responded their neighbors felt they should not fertilize (coded 5), the probability was .52. Second, a number of the technical knowledge/belief variables were significant. Individuals who believed their storm water was treated were much more likely to fertilize (probability of .88) compared with those who did not believe it was treated (probability of .73). Moreover, beliefs on the positive and negative effects of fertilization had large substantive effects. Those who believed pollution was more likely from fertilization (negative effect) were less likely to fertilize. Substantively the probability of fertilization was .89 for those who felt pollution was unlikely (coded 1) compared with .49 for those who felt pollution was likely (coded 5). Likewise, those who felt fertilization did not improve the attractiveness of the lawn were less likely to fertilize (probability of .50) compared with those who believed fertilization would lead to an attractive lawn (probability of .79). Finally, one’s perceived ability to fertilize is a strong predictor on fertilization. The probability of fertilization for those at the extreme difficult end of the scale is .04. In contrast, at the opposite extreme (those who feel it is easy), the probability is .85.

Now let us turn to the affect items (motivations and concerns). In general, these items, as measured, do not seem to have a large connection to fertilization preferences. The motivation of being “a good neighbor” is somewhat significant and has a moderate substantive effect (probability of .51 for those least motivated compared with .82 for those most motivated). However, beyond this item, no other affective items appear to play a statistically significant role.

Model 1 of Table 3 has provided some valuable insight into variables influencing the act of fertilization in general. Now, we want to go beyond this and look at the frequency of fertilization. For that, we turn to Model 2 of Table 3.

Interestingly, many of the variables significant for simply the act of fertilizing are also significant for the frequency of fertilization. For the structural factors, two of the three variables are significant. First, individuals living in Highland Park had a lower predicted frequency of fertilization (2.24 compared with 2.5 in Lino Lakes). Moreover, users of professional lawn services had a higher predicted frequency of fertilization (3.43) compared with homeowner managed (2.4). As for demographics, the only significant variable is the variable for gender, with women having a lower predicted fertilization frequency (2.06) compared with men (2.24).

Turning to the suite of belief variables, we see a much smaller set of variables play a role compared with simply fertilizing. Of the cognitive items, the link between fertilization and an attractive lawn was highly significant. Individuals who believed in the link between fertilization and an attractive lawn had a predicted frequency of 2.46 compared with 1.49 for those who did not make the link. The other cognitive item was the ability variable. At the most difficult end of the scale, the predicted frequency is 1.22 and increased to 2.36 at the easy end of the scale. As for the affective items, two of the motivations play a significant role (“attractive property” and “activity I enjoy”). For maintaining an attractive property, the predicted frequency increases from 1.78 for those who are not motivated by this item to 2.36 for those who are motivated. Likewise, the predicted frequency increases from 2.05 (least supportive) to 2.40 (most supportive) for maintaining the lawn “as an activity I enjoy.”

Finally, let us consider model 3 (high fertilizers). Again, we see a consistent set of variables associated with the “high” fertilizer individuals as well. Of the structural variables, we again see significant differences between suburban Lino Lakes (probability of being a high fertilizer is .07) and urban Highland Park (probability is .03). More substantially, the use of a lawn care company increases the probability of being a “high fertilizer” from .03 to .19. As for sociodemographics, only education is statistically significant, but at the p < .10 level, and the substantive effect is a 3-point difference between the lowest and highest education levels. Finally, the belief variables also show some significant effects. The norm variable is slightly significant and leads to a 5-point increase in the probability of being a “high fertilizer” when going from the lowest to the highest categories of this variable. Similarly, the connection between fertilizer and an attractive lawn leads to about a 7-point increase in the probability. The motivation for an attractive property is slightly significant and leads to about a 4-point increase in the probability. Finally, the ability item is highly significant and leads to almost a 6-point increase in the probability of being a “high fertilizer.”

Robustness Checks

We conducted a variety of additional tests to ensure the robustness of our results. First of all, because the multivariate models lead to a substantial drop in observations, we reestimated the models in Table 3 with multiple-imputation models (MI). We used STATA 12’s chained imputation procedure (a sequential approach) for all observations below 90% complete. This led to imputation of the following variables: (a) length of home ownership, (b) neighbors think I should not fertilize, (c) concerns with local waters, and (d) ease of fertilization. The variables used to predict the imputed variables included all the independent and dependent variables in Table 3. Fortunately, the results from the multiple-imputation models were nearly identical to those in Table 3 so we feel confident that the drop in N is not impacting our results.

Moreover, while we ran Model 2 (frequency) as a Poisson model, a negative binomial model was attempted but the alpha was close enough to zero to indicate that a Poisson model was appropriate and the results were nearly identical. OLS was also considered and provided very similar results to the Poisson model. Finally, given the possible selection bias of only using respondents who did fertilize, we attempted a Heckman selection model but the rho value indicated we did not have a problem with selection bias and the results were similar.

In addition to the results displayed in Models 1 and 2 of Table 3, our online appendix (at http://eab.sagepub.com/supplemental) presents similar models for a set of questions asking about the respondent’s typical fertilization (not year specific). Results from those models are quite similar to the ones presented in Table 3.

Finally, for fertilization frequency, only eight observations fertilized greater than seven times. We tested the impact of these outliers a number of different ways and found no substantial impact. Moreover, results were similar when attempting to correct for the outliers.

Discussion

In many regards, the traditional lawn has become problematic (e.g., Robbins et al., 2001; Robbins & Sharp, 2003a) and attention has focused on how to minimize the problems of the traditional lawn. Some research has ventured into alternatives to the residential lawn (e.g., Larson et al., 2010; Yabiku, Casagrande, & Farley-Metzger, 2008). Other research has attempted to explore how to make the traditional lawn less problematic by focusing on yard care practices (see Cook et al., 2012, for a thorough discussion). One specific area within this research has been focused on fertilization practices. Over the last year alone, there have been numerous studies aimed specifically at fertilization practices (e.g., Blaine et al., 2012; Carrico et al., 2013). Our work complements these existing studies by (a) analyzing a model that combines many of these different hypotheses and conclusions and looks for the most important influences on lawn fertilization behavior across a wide range of plausible factors, (b) exploring the influences on fertilization frequency as well as simply fertilizing, and (c) considering a new geographic context (the Minneapolis/Saint Paul Metropolitan area), along with furthering the discussion around community type and lawn management behavior.

First, our research considered a wide range of variables and demonstrated that fertilization practices are driven by a combination of structural aspects, sociodemographics, informal norms, beliefs, motivations, and perceived ability. To some degree, many of these conclusions have been established before (e.g., Blaine et al., 2012; Carrico et al., 2013; Osmond & Platt, 2000; Robbins & Sharp, 2003a; Templeton et al., 1999) but we demonstrate the robustness of these results by controlling for a wide range of plausible influences. In essence, by including so many theoretically relevant influences and controlling for so many factors, we are able to provide a more refined picture of the influences on fertilization behavior.

What is quite interesting is that there is not one major factor driving fertilization behavior, which also means there is no one magic bullet to fix potential environmental problems that might accompany the use of fertilizer. However, as lawn fertilization is a substantial area where individuals can reduce their N contribution (Fissore et al., 2011), we should consider factors that might be more suitable to help change fertilization practices. This leads us to recommending future research on the role of beliefs, especially the cognitive aspects of beliefs. Our analysis has shown that beliefs are a statistically significant influence on fertilization practices and have a large effect on the probability of fertilization and the frequency of fertilization. They are also the factors that could be directly targeted through education. Certainly, knowledge gaps exist regarding factors surrounding fertilization and yard management in general (e.g., Dahmus & Nelson, 2014; Nielson & Smith, 2005). For example, Nielson and Smith (2005) asked respondents about where the water from the storm drain goes. Fifty-seven percent claimed to not know compared with only 15% who correctly identified that the water went to the nearest steam. In our sample, 44% correctly answered that storm water is not treated, but that still leaves the majority with misinformation. Information campaigns aimed at the pollution problems associated with fertilization along with best practices for fertilization may be successful at changing some practices in a number of ways. Such campaigns may be successful at aiding homeowners to conceive of their yard as cycles that are interconnected to the surrounding water bodies, as opposed to the more general viewpoint of the yard as separate from the surrounding area (Dahmus & Nelson, 2014). Moreover, homeowners may see their yard care practices in a rational cost/benefit manner. Providing additional information about potential externality costs for local water bodies will aid homeowners in making decisions in a more fully informed manner. Again, these factors are just a few of many different factors influencing fertilization practices so we should not expect dramatic swings in fertilization practices but even a modest change in practices may lead to changing norms, which could lead to additional change in practices.

What is also interesting is that the more cognitive components of beliefs and attitudes were consistently associated with fertilizer usage, where the more affective components failed to show consistent associations. Specifically, the concern for local waters variable never reached statistical significance. While this is a common result among research (Askew & McGuirk, 2004; Templeton et al., 1999), we wonder how much of this lack of significance is also tied to a lack of knowledge and education. Substantial research has noted the importance of education in environmental behavior (Byrne & Grewal, 2008; Hines, Hungerford, & Tomera, 1987; Nielson & Smith, 2005), and future research should consider if these affective components require some knowledge-based link to activate a behavior change.

Second, our research demonstrates that many of the factors that drive fertilization in the first place also drive frequency. In general, scholarship has focused more on the act of fertilizing rather than the frequency of the behavior. However, fertilizing and fertilization frequency are two very different aspects (e.g., two people can both fertilize, but one person can fertilize his or her lawn five times a year and the other one time a year). In general, it is largely agreed that some fertilization is beneficial to lawn care and can have positive effects (e.g., Rosen et al., 2006, provide a lengthy discussion concerning fertilization recommendation depending on numerous yard factors). What is of greater concern is overfertilization because this can create excess nutrient runoff, which can contaminate local water bodies. Moreover, as demonstrated by Fissore et al. (2011), the frequency of fertilization is quite variable and highly skewed among households. Essentially, a small portion of total households contribute disproportionately to total fertilization and we should consider what is linked to this behavior at the high end of the curve. With these concerns in mind, our analysis considers frequency of fertilization and those considered “high fertilizers.” We find that many variables linked to frequency are the same as simply fertilizing. This is not without some important exceptions, however. First, the norm variable appears to play a significant role in influencing an individual to fertilize his or her lawn. However, it does not appear to play a much of a role in how frequently this individual fertilizes. Moreover, knowledge regarding the effects of fertilizer and pollution appear to drive the act of fertilizing but not frequency, where knowledge on the effects of fertilizer and an attractive lawn play a much stronger role on fertilizer frequency. It is interesting how the type of knowledge (positive or negative effect) can be linked to different aspects of fertilization.

Finally, we note a considerable increase in fertilization and fertilization frequency in the more suburban Lino Lakes as compared with the more urban Highland Park. We ran a number of comparisons between the two communities to look for connections to these differences (analysis not shown) and while we come up with some insights, we did not find any dominant reason. Both communities differ in a wide range of factors, notably in their beliefs and motivations, factors that are correlated with fertilization. They also were similar in many regards, notably with use of a lawn care company. However, even while controlling for the wide variety of factors in our models, individuals in these communities still showed significant differences in fertilization behavior. This implies that there are additional factors we are not able to capture in our models (e.g., some informal/formal norms, different mindsets/ideologies). Surely, we have only been able to scratch the surface of this relationship and will need to explore this more completely with future research.

We believe this study provides a number of opportunities for future research. Specifically, future research should further study the idea of beliefs and knowledge within the context of the social conversation of neighborhoods because these were shown to be highly significant and substantively important. It would be important to uncover how much behavioral change could be achieved through education, or even what information/knowledge can bring about behavioral change. Furthermore, our focus is on fertilization, but fertilizing is just one of a number of yard management behaviors and one piece of the overall yard nutrient cycle and management. Future studies should consider the suite of yard management behaviors and their impacts on each other. Finally, more research in varied locations would allow for a fuller understanding of fertilization practices and yard management in general. Currently, each study has a variety of hypotheses and results. Some of these overlap and others do not. By broadening the geographic study area, we are able to find which hypotheses, and accompanying results, are generalizable across areas and which are more geographically specific. While research has expanded substantially and the number of sites has grown, more is certainly needed.

Certainly in every study there are some limitations, and ours is no exception. First, we only study one geographic location (Minneapolis/Saint Paul, Minnesota Metropolitan Area), and most respondents were white and older than average. It is possible that neighborhoods with different ethnic/racial or age makeups would yield different results. While the limited scope somewhat impacts the overall generalizability of our results, our results align with expectations and findings in other locations as well. Second, while we have solid measures on whether the individual fertilized and their noted frequency, we lack an understanding of how much fertilizer is being used at each application and how much of the lawn is being fertilized per application. Future studies should probe further for more detailed information on exact fertilization practices. Moreover, all of the data obtained are self-reported from the respondent. While self-reported survey information is a common source of data, we acknowledge that incorrect information could be provided. Specifically, with fertilization frequency, it is likely that many individuals do not keep track of all of their fertilization applications so their responses are estimates. At the same time, fertilization is likely an easy behavior for individuals to track—without obvious reasons for systematic over- or underreporting. The greater precision gained with more costly studies may not yield further insight. Finally, we did not consider the homeowner knowledge about their soil type and slope. Homeowners may have some sense of the runoff potential from their lawns and that this might influence lawn management behaviors.

In the end, these results provide a more complete understanding of what drives the two behaviors—fertilization and fertilization frequency. We considered an extensive list of factors and found that these fertilization behaviors are linked to a combination of structural aspects, sociodemographics, informal norms, beliefs, motivations, and perceived ability. Most notably among these factors were the many influences revolving individual-level beliefs and attitudes. Moreover, while these two behaviors have many of the same influences, there are some differences, again the most notable being within beliefs and attitudes. This common theme of beliefs and attitudes being the most notable influence is important for researchers to consider. Fertilizing one’s lawn is not an automatic behavior but based on a decision-making process. Some of the factors leading to this decision are less controllable from a researcher and practitioner standpoint (i.e., structural and sociodemographics), but attitudes and beliefs surrounding fertilization are prime areas for intervention. This fits in with the broader literature of proenvironmental behavior being linked to proenvironmental beliefs/knowledge. Beliefs and knowledge play a significant role in driving behavior and if we can understand how to influence beliefs, we may be able to shape a variety of yard management behaviors.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Science Foundation, Division of Behavioral and Cognitive Sciences (Award Number 0908998).

Author Biographies

Nicholas F. Martini is a postdoctoral research associate in the Department of Forest Resources at the University of Minnesota. His research focuses on attitudes, beliefs, and values, and how they interact with a wide range of behaviors, including the urban environment.

Kristen C. Nelson is a professor in the Departments of Forest Resources and Fisheries, Wildlife, and Conservation Biology at the University of Minnesota. Her urban ecosystems research focuses on an interdisciplinary understanding of what influences homeowner decisions (attitudes, norms, perceived control, sociodemographic factors), public land manager vegetation decision making, how behaviors influence biogeochemistry/environmental services/biodiversity, and social change in complex systems (social networks).

Sarah E. Hobbie is a professor in the Department of Ecology, Evolution and Behavior at the University of Minnesota. Her research focuses on three main areas: (a) the influence of changes in atmospheric composition and climate on ecosystem processes, (b) the effects of urbanization and suburbanization on biogeochemical cycles, and (c) the influence of plant species on biogeochemical processes.

Lawrence A. Baker is research professor in the Department of Bioproducts and Biosystems Engineering at the University of Minnesota. His research focuses on the integration of ecological thinking into urban sustainability, the development of a translational research framework for urban sustainability, and a development of a vision of adaptive cities.

References

Ajzen

(1991). The theory of planned behavior. Organizational behavior and human decision processes, 50, 179-211.

Ajzen

(2002). Perceived behavioral control, self-efficacy, locus of control, and the theory of planned behavior. Journal of Applied Social Psychology, 32, 665-683.

Askew

McGuirk

(2004). Watering the suburbs: Distinction, conformity and the suburban garden. Australian Geographer, 35, 17-37.

Bandaranayake

Qian

Y. L.

Parton

W. J.

Ojima

D. S.

Follett

R. F.

(2003). Estimation of soil organic carbon changes in turfgrass systems using the CENTURY model. Agronomy Journal, 95, 558-563.

Beard

J. B.

Green

R. L.

(1994). The role of turfgrasses in environmental protection and their benefits to humans. Journal of Environmental Quality, 23, 452-460.

Blaine

T. W.

Clayton

Robbins

Grewal

P. S.

(2012). Homeowner attitudes and practices towards residential landscape management in Ohio, USA. Environmental Management, 50, 257-271.

Butterfield

B. W.

(2004). 2004 Environmental lawn and garden survey. South Burlington, VT: National Gardening Association.

Byrne

L. B.

Grewal

(2008). Introduction to ecological landscaping: A holistic description and framework to guide the study and management of urban landscape parcels. Cities and the Environment, 1(2), Article 3.

Carpenter

S. R.

Caraco

N. F.

Correll

D. L.

Howarth

R. W.

Sharpley

A. N.

Smith

V. H.

(1998). Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecological Applications, 8, 559-568.

10.

Carrico

A. R.

Fraser

Bazuin

J. T.

(2013). Green with envy: Psychological and social predictors of lawn fertilizer application. Environment & Behavior, 45, 427-454.

11.

Chowdhury

R. R.

Larson

Grove

Polsky

Cook

(2011). A multi-scalar approach to theorizing socio-ecological dynamics of urban residential landscapes. Cities and the Environment, 4(1), Article 6.

12.

Cook

E. M.

Hall

S. J.

Larson

K. L.

(2012). Residential landscapes as social-ecological systems: A synthesis of multi-scalar interactions between people and their home environment. Urban Ecosystems, 15, 19-52.

13.

Dahmus

M. E.

Nelson

K. C.

(2014). Yard stories: Examining residents’ conceptions of their yards as part of the urban ecosystem. Urban Ecosystems, 17(1), 173-194. doi:10.1007/s11252-013-0306-3

14.

Dunlap

R. E.

Jones

R. E.

(2002). Environmental concern: Conceptual and measurement issues. In Dunlap

R. R.

Michelson

(Eds.), Handbook of environmental sociology (pp. 482-524). Wesport, CT: Greenwood.

15.

Feagan

R. B.

Ripmeester

(1999). Contesting natural(ized) lawns: A geography of private green space in the Niagara region. Urban Geography, 20, 617-634.

16.

Fishbein

Ajzen

(1975). Belief, attitude, intention and behavior: An introduction to theory and research. Reading, MA: Addison-Wesley.

17.

Fissore

Baker

L. A.

Hobbie

S. E.

King

J. Y.

McFadden

J. P.

Nelson

K. C.

Jakobsdottir

(2011). Carbon, nitrogen, phosphorus fluxes in household ecosystems in the Minneapolis-Saint Paul, Minnesota urban region. Ecological Applications, 21, 619-639.

18.

Greenlee

A. R.

Ellis

T. M.

Berg

R. L.

(2004). Low-dose agrochemicals and lawn-care pesticides induce developmental toxicity in murine preimplantation embryos. Environmental Health Perspectives, 112, 703-709.

19.

Hines

J. M.

Hungerford

H. R.

Tomera

A. N.

(1987). Analysis and synthesis of research on responsible environmental behavior: A meta-analysis. The Journal of Environmental Education, 18, 1-8.

20.

Jenerette

G. D.

Harlan

S. L.

Brazel

A. J.

Jones

Larsen

Stefanov

W. L.

(2007). Regional relationships between vegetation, surface temperature, and human settlement in a rapidly urbanizing ecosystem. Landscape Ecology, 22, 353-365.

21.

Jenkins

V. S.

(1994). The lawn: A history of American obsession. Washington, DC: Smithsonian Institution.

22.

Larson

K. L.

Casagrande

Harlan

S. L.

Yabiku

S. T.

(2009). Residents’ yard choices and rationales in a desert city: Social priorities, ecological impacts, and decision tradeoffs. Environmental Management, 44, 921-937.

23.

Larson

K. L.

Cook

Strawhacker

Hall

S. J.

(2010). The influence of diverse values, ecological structure, and geographic context on residents’ multifaceted landscaping decisions. Human Ecology, 38, 747-761.

24.

Larson

K. L.

Ibes

D. C.

White

D. D.

(2011). Gendered perspectives about water risks and policy strategies: A tripartite conceptual approach. Environment & Behavior, 43, 415-438.

25.

Law

N. L.

Band

L. E.

Grove

J. M.

(2004). Nitrogen input from residential lawn care practices in suburban watersheds in Baltimore County, MD. Journal of Environmental Planning and Management, 47, 737-755.

26.

Lubowski

R. N.

Vesterby

Bucholtz

Baez

Roberts

M. J.

(2006). Major uses of land in the United States, 2002 (Economic Information Bulletin Number 14). United States Department of Agriculture. Retrieved from http://www.ers.usda.gov/publications/eib-economic-information-bulletin/eib14.aspx#.UZ93n5w1B8E

27.

Milesi

Running

S. W.

Elvidge

C. D.

Dietz

J. B.

Tuttle

B. T.

Nemani

R. R.

(2005). Mapping and modeling the biogeochemical cycling of turf grasses in the United States. Environmental Management, 36, 426-438.

28.

Nassauer

J. I.

Wang

Dayrell

(2009). What will the neighbors think? Cultural norms and ecological design. Landscape and Urban Planning, 92, 282-292.

29.

Nelson

K. C.

Dahmus

Woodside

(2011). Our yard care choices in urban living survey. St. Paul: University of Minnesota. Retrieved from http://www.tchep.umn.edu/

30.

Nelson

K. C.

Grayzeck

King

Hobbie

Baker

McFadden

(2008). Our household choices in urban living survey. St. Paul: University of Minnesota. Retrieved from http://www.tchep.umn.edu/

31.

Nielson

Smith

C. L.

(2005). Influences on residential yard care and water quality: Tualatin watershed, Oregon. Journal of the American Water Resources Association, 41, 93-106.

32.

Osmond

D. L.

Hardy

D. H.

(2004). Characterization of turf practices in five North Carolina communities. Journal of Environmental Quality, 33, 565-575.

33.

Osmond

D. L.

Platt

J. L.

(2000). Characterization of suburban nitrogen fertilizer and water use on residential turf in Cary, North Carolina. HortTechnology, 10, 320-325.

34.

Robbins

Birkenholtz

(2003). Turfgrass revolution: Measuring the expansion of the American lawn. Land Use Policy, 20, 181-194.

35.

Robbins

Polderman

Birkenholtz

(2001). Lawns and toxins: An ecology of the city. Cities, 18, 369-380.

36.

Robbins

Sharp

(2003a). Producing and consuming chemicals: The moral economy of the American lawn. Economic Geography, 79, 425-451.

37.

Robbins

Sharp

(2003b). The lawn-chemical economy and its discontents. Antipode, 35, 955-979.

38.

Rosen

C. J.

Horgan

B. P.

Mugaas

R. J.

(2006). Fertilizing lawns (FO-03338). University of Minnesota extension. Retrieved from http://www.extension.umn.edu/distribution/horticulture/DG3338.html

39.

Stern

P. C.

Dietz

Abel

Guagnano

G. A.

Kalof

(1999). A value-belief-norm theory of support for social movements: The case of environmentalism. Human Ecology Review, 6, 81-97.

40.

Templeton

S. R.

Yoo

S. J.

Zilberman

(1999). An economic analysis of yard care and synthetic chemical use: The case of San Francisco. Environmental & Resource Economics, 14, 385-397.

41.

Yabiku

S. T.

Casagrande

D. G.

Farley-Metzger

(2008). Preferences for landscape choice in a southwestern desert city. Environment & Behavior, 40, 382-400.

42.

Zhou

Troy

Grove

(2008). Modeling residential lawn fertilization practices: Integrating high resolution remote sensing with socioeconomic data. Environmental Management, 41, 742-752.

43.

Zhou

Troy

Grove

J. M.

Jenkins

J. C.

(2009). Can money buy green? Demographic and socioeconomic predictors of lawn-care expenditures and lawn greenness in urban residential areas. Society & Natural Resources, 22, 744-760.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.41 MB