Elsevier

Landscape and Urban Planning

Volume 165, September 2017, Pages 22-29
Landscape and Urban Planning

Research Paper
Legacy effects and landscape choices in a desert city

https://doi.org/10.1016/j.landurbplan.2017.04.014Get rights and content

Highlights

  • We found that legacy effects influence landscape choices and preferences in Phoenix, Arizona.

  • Long-time Phoenicians chose grass yards due to socialization to local customs.

  • Findings are contrary to older studies and common assumptions about newcomers’ lawn demands.

  • Residents in older neighborhoods both prefer and have grass landscapes compared to others.

  • Legacy effects can impede landscape change in cities due to established traditions and norms.

Abstract

Lawns are a traditional landscape choice in urban neighborhoods, and their pervasiveness results in extensive irrigation around the world. Focusing on the semi-arid metropolis of Phoenix, Arizona, this paper examines residents’ actual and preferred landscapes for both front and back yards; specifically considering grass (mesic) yards, gravel-based (xeric) yards, and mixed (oasis) yards that include some grass and some gravel. Since xeric yards can conserve water, practitioners have promoted them as a drought-tolerant alternative to traditional lawns. Understanding the factors that motivate or constrain landscaping choices can facilitate the transition from lawns toward more naturalistic alternatives. This paper provides an original analysis of how legacies in the Phoenix region—including local landscape traditions and development history—affect yard choices. Using survey data and inferential statistics, we found that longer-term residents—as measured by the proportion of a resident’s lifetime spent in the Phoenix region—more often chose grassy landscapes compared to newcomers. This is counter to the common assumption that newcomers to the desert prefer lawns; instead, long-time residents seem to be accustomed to the long-established luxuriant landscapes of ‘the Phoenix Oasis.’ Residents of older neighborhoods also chose grassier landscaping compared to residents in newer areas, who tended to choose xeric yards. Altogether, these findings reflect the lasting legacies that previous landscaping choices have in urban environments, where changes in preferences and practices take time. Ultimately, the legacy effects of past choices often persist, thereby impeding efforts to promote drought-tolerant and naturalistic landscapes.

Introduction

Lawns are highly desired landscapes across the United States, with over 10–16 million hectares of land covered by turfgrass (Milesi et al., 2005). This estimate includes (but is not limited to) residential neighborhoods, where grass tends to be the traditional landscape type in yards. In desert cities such as Phoenix, Arizona, irrigating lawns contribute significantly to water consumption (Wentz & Gober, 2007). Due to the high water requirements of grass, coupled with reoccurring droughts and potential water scarcity, scholars, practitioners, and conservation advocates have pushed for replacing the traditional lawn with alternative land-cover types or otherwise reducing the amount of grass in residential neighborhoods (Larson, Casagrande, Harlan, & Yabiku, 2009; Robbins & Birkenholtz, 2003). In order to shift urban landscaping toward low water-use and climate-adapted yards, knowledge about what motivates and constrains landscape choices is imperative.

The desire for lawns has been driven by several factors including their appearance, familiarity, and feelings of social obligation that are largely driven by social norms (Larson and Brumand, 2014, Larson et al., 2009, Robbins, 2007). In the arid western U.S., a common expectation regarding lawns is that people migrating from relatively humid regions such as the Midwest have created and reinforced the demand for traditional grass landscapes (McPherson & Haip, 1989). However, several studies conducted in the metropolitan Phoenix area have found the opposite trend—that is, longer-term residents tend to prefer lawns more so than residents who are relatively new to the desert region. Although three different studies have supported this tendency (Larson et al., 2009; Martin, Peterson, & Stabler, 2003; Yabiku, Casagrande, & Farley-Metzger, 2008), thereby strengthening its reliability and generalizability in Phoenix, the variables and samples examined were limited in each of these studies. We therefore build upon this work by analyzing a more robust array of landscape choices (i.e., preferred and actual landscape types for front and back yards) in relation to tenure of residency as well as the age of housing, both of which capture legacy effects in terms of the yard choices made by long-term residents who are accustomed to local norms as well as the landscaping decisions embedded in neighborhoods built during different times periods.

Researchers have studied the legacies of past decision-making, for example, demonstrating the effects of historic land uses (e.g., agriculture) on present-day ecological conditions such as biogeochemistry (Grimm et al., 2008). Although several urban ecological studies have examined legacy effects (e.g., Clarke, Jenerette, & Davila, 2013; Hall et al., 2013, Hope et al., 2003; Lewis, Kaye, Gries, Kinzig, & Redman, 2006; Lubbe, Siebert, & Cilliers, 2010), less social science research has been conducted on the legacy effects in cities. A review of literature on the urban ecology of residential landscapes found relatively few studies on legacy effects, emphasizing a need for more research in this area (Cook, Hall, & Larson, 2012). We address this gap by examining how the history of development (as indicated by age of housing) and residency in the region (specifically as measured by the portion of one’s life spent in the study region) affect actual and preferred landscape types in people’s front and back yards. Employing inferential statistics with quantitative data from a questionnaire, we also examine the relationship between landscape choices and both age and years of residency in metropolitan Phoenix, which together were used to calculate the portion of a resident’s life spent in the area.

To clarify the terminology in this paper, a yard encompasses the outdoor areas of homes in single-family residential parcels, which are often split into public front yards and relatively private back yards (Larsen & Harlan, 2006). Residential yards can encompass a variety of landscape components inclusive of living (e.g., trees, plants) and non-living (e.g., gravel, mulch, or cement) elements. Traditionally, single-family residential yards in the US and beyond have been dominated by grass cover, also known as lawn. In diverse regions of the world, people have called for yards that are more water-tolerant, biodiverse, and otherwise lessen the negative impacts of landscaping choices on the environment (Cook et al., 2012; Goddard, Dougill, & Benton, 2010; Lubbe et al., 2010).

Although residential landscapes can be described in a variety of ways, we adapted a commonly used typology for desert cities (e.g., by Larson et al., 2009, Martin et al., 2003, Yabiku et al., 2008). In particular, scholars have contrasted lawn-dominated yards (known as mesic, literally wet, because of their dependence on irrigation) with yards that have no or partial grass. Given the importance of water conservation in arid cities, xeric (dry) landscapes have been described as a drought-tolerant and desert-like alternatives to mesic lawns. In our typology, xeric yards entail landscapes that encompass rock or gravel groundcover with low water-use plants (Martin et al., 2003). Yet not all yards are comprised of all or no grass; some yards encompass some grass and some gravel groundcover, which are called oasis landscapes to reflect some verdant elements in the midst of an otherwise arid landscapes. At the regional scale, the Phoenix area—also known as the Valley of the Sun—has been described as an oasis since its landscapes provide a relatively lush refuge compared to the native Sonoran Desert ecosystem within which it is situated.

In the next sections, we present the research that informs this study while establishing the theoretical expectations between landscape choices and tenure of residency as well as age of development. We then discuss the methods of data collection and analyses before presenting and discussing our results.

As noted in the introduction, a few studies seem to suggest that the ‘oasis mentality’ may be entrenched in the landscape norms and choices of established Phoenicians, since research has shown that long-time Phoenix residents tend to choose grass yards more so than relative newcomers (Larson et al., 2009, Martin et al., 2003, Yabiku et al., 2008). Before detailing the findings of this comparatively new research, it is important to note that earlier studies have indicated the opposite effect. For example, McPherson and Haip (1989) noted that throughout the late-1800s to mid-1900s, “people with their Eastern and Midwestern landscape images and values” (440) were responsible for bringing exotic trees and turfgrass species (e.g., Bermuda and winter rye grass) to the southwestern U.S. Zube at al. (1986: 8) concurred: “The grass lawn tradition dates from the earliest period of Anglo settlement in both Tucson and Phoenix when immigrants from more temperate climates moved to these desert locations and created the greener landscape they had left behind in the Midwest and Southeast.” Kennedy and Zube (1991) empirically demonstrated this expectation in a study of Tucson, Arizona, which found long-time residents of the city (which is located just 185 km, or 115 miles, to the south of Phoenix) preferred native desert plants and low water-use vegetation more so than newcomers.

Among the more recent Phoenix-based studies, the earliest one (Martin et al., 2003) showed that long-term residents of Phoenix tend to prefer grassy landscapes while those native to Arizona prefer mesic landscapes more than relatively xeric alternatives. More specifically, native Arizonans least preferred xeric landscapes (16%) compared to those from other regions, with xeric yard preferences ranging from a low of 21% for residents from the Southwest to a high of 43% among those from the Great Plains. Another study by Larson et al. (2009) found similar trends as Martin et al. (2003) in that native Phoenicians more often preferred and actually had mesic back yards compared to newer Phoenix residents. The trend for actual landscapes was similar but less pronounced for front yards; however, preferences for front yards were similar across native and non-native residents. Although informative, neither of these studies conducted analyses to show whether or not these differences were statistically significant.

Yabiku et al. (2008) hypothesized and found that years spent in the Phoenix region had a statistically significant, negative effect on preferences for desert-like xeric landscapes. These researchers purported that “socialization”—defined as “the process through which individuals learn how to live within society or within a specific group” (386)—affects landscape choices. While socialization is a complex process, Yabiku et al. showed how familiarity with local landscapes—as measured by longer residency in the region—results in disdain for xeric landscapes. This result was also reported by a qualitative study that found ‘people get sick of the desert’ (Larson et al., 2009).

Regardless, the sample in Yabiku et al.’s (2008) study was limited to a narrow demographic of mostly young adults—many with small children—who lived in a residential village of a university campus. This may be the reason why the majority of their sample preferred grass yards—that is, since these are safer environments for children compared to the rock-covered alternatives that often have spiny cactuses and plants (Larson et al., 2009). In contrast to Yabiku et al.’s narrow sample, another survey of a relatively diverse sample of residents from the Phoenix area exhibited more varied preferences, wherein approximately one-third preferred each mesic, oasis, and xeric (Larsen & Harlan, 2006).

Considering length of residency in a region, local customs must be considered. In the case of Phoenix, the region has historically been marketed as a luxuriant oasis, with promotional campaigns such as “the desert is a myth” (Larson et al., 2009). Thus, although the region has seen a rise in xeric, drought-tolerant landscapes, earlier research has shown that Phoenix residents prefer irrigated grass more so than Tucson residents, who instead held stronger preferences for native desert landscapes (Zube, Simcox, & Law, 1986). Whether this distinction still exists is beyond the scope of this paper (since we focus solely on trends in metropolitan Phoenix); however, we will return to this point later on in the discussion, as the history and geography of particular regions are important considerations when examining social legacies in landscaping choices and urban ecology.

Research on legacy effects has shown that previous land-use and land-cover decisions have the potential to significantly impact social and ecological conditions and dynamics into the future (Cook et al., 2012). Past decisions continue to affect current ecological conditions due to lasting legacies involved with the application of chemicals—such as fertilizers used in agriculture—that persist in the environment and thereby affect biogeochemistry (Grimm et al., 2008, Liu et al., 2007). In addition, decisions about landscape structure—including vegetation composition—can have lasting effects (Larson et al., 2010; Robbins, 2007 Robbins, 2007). For example, the original landscapes installed by residential developers or homeowners may persist long into the future, as people may not prioritize (perhaps cannot afford) changing their yards (Larsen & Harlan, 2006).

The inheritance of conventional grass yards results in what Robbins (2007) calls the ‘lawn subject,’ as residents become beholden to managing their existing landscapes according to preexisting standards and social norms. Applied to traditional grass yards, or the “Industrial Lawn” standard (Bormann, Balmori, & Geballe, 2001), people who inherit lawns will manage them according to the hyper-green, weed-free standard established by turfgrass specialists and the industry that supports them. Of course, this is accomplished through irrigation, fertilization, pesticide use, and other maintenance decisions that have short-term and long-term implications on the environment and society.

Although lawns have been the dominant norm in residential neighborhoods since the early 1900s, landscape choices have started to shift toward non-lawn alternatives (Larson et al., 2009). In the Southwestern U.S. specifically, desert-like (xeric) landscapes became increasingly popular post-World War II (Kennedy & Zube, 1991) due to growing appreciation for natural landscapes, changing leisure preferences away from gardening, and increasing concerns about water scarcity (Hecht, 1975, McPherson and Haip, 1989). The availability of plants in nurseries—which has changed over time—also significantly shapes vegetation choices and composition. In a study of Los Angeles, for example, Pincetl, Prabhu, Gillespie, Jenerette, and Pataki (2013) found that the number of tree species offered at nurseries increased over time, especially since the 1990s. Though this California study did not identify a significant increase in native trees over time, other trends—such as the publication of low water-use plant guides and the establishment of the Arizona Native Plant Society—demonstrate a rising interest in desert-adapted and local vegetation in the 1970s (McPherson & Haip, 1989).

In summary, since landscaping traditions have changed over time, the age of housing in neighborhoods could have a lasting effect on residential landscapes (Larsen & Harlan, 2006). Following from historic trends, older homes are likely to have more grass whereas newer homes are likely to have less grass, and in the case of Phoenix, Arizona, more xeric-style yards. Indeed, age of housing has been found to affect urban ecological characteristics including vegetation and biodiversity in other studies (Hope et al., 2003). A study in Baltimore, Maryland revealed that the amount of vegetation—both in terms of grass and trees—is higher in older properties than newer ones (Grove, Troy, O’Neil-Dunne, Cadenasso, & Pickett, 2006; Troy, Grove, O’Neil-Dunne, Pickett, & Cadenasso, 2007). Studies in both Los Angeles (Clarke et al., 2013) and Phoenix (Ripplinger, Franklin, & Collins, 2016) have also found higher biodiversity (specifically for trees and plants, respectively) in older homes compared to new ones.

Based on a review of relevant literature, we now lay out the expected relationships between landscapes choices and our explanatory variables. Regarding age of housing, we expect that residents who live in older neighborhoods will choose traditional grass (mesic) landscapes more so than those in newer areas, where landscape choices will tend to be xeric or oasis. Mesic yards will likely be more common in actual yards compared to expressed preferences given the lasting effects of past decisions as well as the fact that residents’ preferences are not always actualized in the landscapes they maintain (Larsen & Harlan, 2006). This is especially the case in visible front yards, where historic traditions are more likely reinforced by social norms (as compared to private back yards).

With respect to residency, we anticipate that people who have lived longer in the Phoenix region—as measured by both years of residence and the portion of a person’s life spent in the Phoenix region—will prefer mesic lawns more so than relative newcomers. This expectation is consistent with past research (Larson et al., 2009, Martin et al., 2003, Yabiku et al., 2008), though validation is needed give the limitations of those studies. Since the portion of a person’s life spent in a particular location is a more accurate indication of how accustomed they are to that place, we anticipate that this variable will be more strongly related to landscape choices compared to the number of years a personal has lived in the region (regardless of age).

Whereas housing age is likely to influence actual landscapes that were originally installed and are now maintained, tenure of residency may influence preferences more so than actual yards since existing landscapes are subject to constraints. This is especially true for front yards, which are more likely limited by social institutions such as neighborhood customs and rules (Larson & Brumand, 2014; Nassauer, Wang, & Dayrell, 2009). In contrast, backyards tend to be landscaped according to personal preferences (Larsen & Harlan, 2006), and thus we expect that residents are more likely to realize their preferences in private areas of the home.

Lastly, although age is not a central variable in the analysis that follows, we assess the relationship between residents’ age and landscaping choices since age is embodied in the tenure of residency variable (i.e., number of years in the region divided by age). We anticipate that older residents will more prefer xeric landscapes for all yard choices, since these alternatives are often seen as requiring less maintenance compared to mesic lawns (Larson et al., 2009).

Section snippets

Survey implementation

A questionnaire was conducted from May 26, 2011 to January 6, 2012 to collect the data analyzed for this paper. The sample represents a stratified random approach wherein forty-five neighborhoods (based on census block groups) were chosen to represent the demographic (e.g., based on age, income, ethnicity) and geographic (e.g., based on location) profile of the Phoenix region. A random sample of approximately 18–20 respondents were surveyed per neighborhood. This resulted in a response rate of

Findings

Before reporting on the relationships that are central to this research, we first present residents’ reported landscape choices and the correlations among them as well as the explanatory variables.

Discussion

This research demonstrates that past decisions persist in modern-day land cover, creating legacy effects in residential landscaping choices. The significance of the results will now be discussed in relation to past research as well as the practical implications for encouraging more naturalistic, drought-tolerant landscapes in regions such as semi-arid Phoenix, Arizona.

Our analysis identified a legacy effect in which residential landscapes from previous historical periods shape present-day urban

Conclusion

This study demonstrated the legacy effects associated with landscape choices in metropolitan Phoenix, Arizona; specifically, people who have lived more than half of their lives in the region tend to prefer and have more grass compared those who have spent less of their lifetime in the region. This finding is contrary to both earlier research and the common expectation that newcomers from relatively humid regions are responsible for the grassy landscape throughout the metro area. The long-time

Acknowledgements

This material is based upon work supported by the National Science Foundation under Grant No. BCS-1026865, Central Arizona-Phoenix Long-Term Ecological Research: Urban Sustainability in the Dynamic Environment of Central Arizona, and Grant No. SES-1462086, Decision Making under Uncertainty, Decision Center for a Desert City III: Transformational Solutions for Urban Water Sustainability Transitions in the Colorado River Basin. Any opinions, findings and conclusions or recommendation expressed in

References (30)

  • N.B. Grimm et al.

    Global change and the ecology of cities

    Science

    (2008)
  • J.M. Grove et al.

    Characterization of households and its implications for the vegetation of urban ecosystems

    Ecosystems

    (2006)
  • S.J. Hall et al.

    Legacies of prehistoric agricultural practices within plant and soil properties across an arid ecosystem

    Ecosystems

    (2013)
  • M.E. Hecht

    The decline of the grass lawn tradition in Tucson

    Landscape

    (1975)
  • D. Hope et al.

    Socioeconomics drive urban plant diversity

    Proceedings of the National Academy of Sciences

    (2003)
  • Cited by (42)

    • Understanding individual and diffusion behaviors related to native plant gardening

      2022, Journal of Environmental Psychology
      Citation Excerpt :

      It is important to consider that this sample was specifically designed to target Fort Collins residents living in single-family homes, so results may differ in other cultural and geographic contexts. Our results may be influenced by United States and Western cultural norms surrounding landscaping and residential land management decisions such as social pressures to maintain a well-manicured lawn (Larson et al., 2017; Robbins, 2007), as well as ideals of individualism and concerns about privacy and private property in the United States. The dry climate of Colorado, and drought conditions in surrounding states across the western US (National Drought Mitigation Center et al., 2021), may also motivate people in Fort Collins, Colorado to take more interest in native plant gardening for its water conservation benefits than people in the Eastern United States or in regions where drought is not as relevant.

    • The environmental consequences of residential land tenure in single family neighborhoods

      2022, Land Use Policy
      Citation Excerpt :

      One strategy for reducing residential demand for irrigation is to replace water intensive turf grass with drought tolerant landscaping. Yet, households rarely completely change landscape type in part because it is a large investment (e.g., between grass, desert, etc.) (Avolio et al., 2018; Robbins and Sharp, 2003; Larson et al., 2017). In the United States, most municipalities do not have strong regulations on the use of turf grass lawns in new or renovated yards, and mandatory retrofits are non-existent (Sisser et al., 2016; Larson et al., 2020).

    • Landscaping preferences influence neighborhood satisfaction and yard management decisions

      2021, Urban Forestry and Urban Greening
      Citation Excerpt :

      Similar to findings in Ostoić et al. (2017), the dissatisfaction of these pro-ecological respondents who preferred mesic landscaping was likely driven by the fact that vegetation fell short of their expectations of a “green” city, reflecting their desire for more trees and shade in their neighborhood. Literature has shown that the longer someone lives in the Phoenix region, the more likely they prefer and have grassy, mesic landscaping (Larson et al., 2017). However, we found that for respondents who lived in Phoenix longer—and who also preferred xeric landscapes—the legacy of the Phoenix Oasis appears to have decreased satisfaction, probably because people with xeric preferences wanted more desert landscaping.

    View all citing articles on Scopus
    View full text