Research ArticleArticles

Long-Term Growth of Highway Rights-of-Way Trees

Allyson B. Salisbury, Jason W. Miesbauer and Andrew K. Koeser
Arboriculture & Urban Forestry (AUF) September 2022, 48 (5) 293-307; DOI: https://doi.org/10.48044/jauf.2022.022

Abstract

Background: Highway rights-of-ways (ROWs, or verges) contain multiple stressors which can influence tree growth, including compacted soils, soils with little topsoil, poor drainage, air and soil pollutants, construction activities, and de-icing salts in cold climates. Yet highway ROWs often provide ample planting space for growing trees, which can contribute to the mitigation of negative environmental impacts associated with highways. Methods: For this study, we assessed the trunk diameter of 1,058 trees from 11-, 22-, and 31-year-old planting cohorts along a highway in the Chicago metropolitan region (Illinois, USA) to examine factors which could influence long-term growth. We analyzed the impact of location factors within the ROW (e.g., distance and elevation relative to highway, slope, and aspect) on trunk diameter at breast height (DBH), since these factors are relevant to the landscape design process. Using estimates from i-Tree, we compared carbon sequestration, carbon storage, runoff reduction, and air-pollution removal within and among the 3 cohorts. Results: Of the 6 site location characteristics we evaluated, no single characteristic consistently impacted DBH, though some characteristics were significant within a single cohort. DBH measurements of most species were smaller than model predictions based on existing urban tree models. Since all cohorts included large- and small-statured trees, and even within species DBH could be highly variable, the range in per-tree ecosystem services varied substantially within cohorts, especially the 31-year-old cohort. Conclusions: These findings highlight both the potential for and challenges of growing trees alongside highways.

Keywords

INTRODUCTION

Urban forests are expected to provide a variety of benefits for their communities while minimizing costs and disservices (Turner-Skoff and Cavender 2019; Roman et al. 2021). However, a range of human and biophysical factors can increase tree mortality (Hilbert et al. 2019) and compromise growth in urban landscapes (Vogt et al. 2015). Observing and modelling the effects of exposure to disturbances and stressors on the structure and function of the urban forest is a necessary step to assess the overall vulnerability of urban forests and take actions to limit potential losses of ecosystem services (Steenberg et al. 2017). While there are multiple ways to evaluate the structure of urban forests (Leff 2016), tree size is a particularly valuable metric. Many urban-forest functions such as building energy-usage reduction, air-pollutant removal, and carbon sequestration directly scale with tree size (Nowak et al. 2008).

Conditions in the built environment often, though not always, lead to reductions in tree growth and consequently the benefits provided by mature trees. Within urban environments, growth is often greater in park or garden settings compared to street trees grown in strips or planting cutouts in close proximity to nonarterial streets (De Lacy and Shackleton 2014; North et al. 2018). The presence of pavement can limit soil moisture, aeration, and nutrient availability (Hodge and Boswell 1993) and increase surface temperatures (Chen et al. 2017), reducing growth. In some cases, open growing conditions found in some urban planting sites favor greater aboveground growth compared to trees in forest settings (Rhoades and Stipes 1999; Smith et al. 2019). Smith et al. (2019) also observed that any potential benefits to carbon sequestration resulting from faster growth of urban trees was offset by shorter life spans. Growth responses to urban settings in the first several years after planting can also vary among species and be influenced by management decisions and practices such as planting season, initial caliper, irrigation, and mulching (Lawrence et al. 2012; Koeser et al. 2014; Vogt et al. 2015). While land-use type can also influence growth (Lawrence et al. 2012), factors that influence tree size within other urban landscapes beyond street trees and residential areas are less studied though no less critical for identifying management interventions to improve growth.

One such urban land-use type that could have the potential to support substantial tree growth is highway rights-of-way (ROW; also referred to as road verges or road reserves). Highways, high-speed and high-volume roadways which can be flanked by strips of undeveloped land, are sources of air and noise pollution, stormwater runoff, and habitat and neighborhood fragmentation (Forman et al. 2002). Trees planted adjacent to highways can help mitigate these negative impacts (Rogers and Evans 2015), provided they can grow to maturity in a reasonable time frame. Unlike street trees which are often planted in limited soil volumes and in close proximity to the road, highway trees are generally planted in much larger soil volumes with open growing conditions and set farther back from the highway to provide clear zones for run-off-road accidents. The highway construction process can significantly alter the land adjacent to the road by topsoil removal and compaction, creating site conditions unfavorable for tree survival and growth (Somerville et al. 2018; McGrath et al. 2020). Highway trees are also exposed to pollution as well as de-icing salts in cold climates (Bryselbout et al. 2000; Fay and Shi 2012). While typically a narrow strip of land, ROW can have highly variable growing conditions, as many soil properties exhibit distance-dependent relationships with the road (Bryson and Barker 2002; Akbar et al. 2012; Werkenthin et al. 2014). Highway ROW can also contain multiple microtopographic features such as drainage ditches, foreslopes, and backslopes (Jimenez et al. 2011; Neher et al. 2013). Identifying the relationship between locations and features in the highway ROW and tree growth could inform decisions about where and what species to plant in a setting that would greatly benefit from trees.

The goal of this study was to examine the impacts of species and cultivar choices and location within the highway ROW on tree size and consequently ecosystem service provision in a highway setting. We used a set of trees planted between 1988 and 2008 along a highway in northern Illinois to investigate the following questions:

  • 1) Do location and site factors in the ROW (e.g., the distance from the highway or location on sloped ground) influence tree growth?

  • 2) How does highway tree size compare to allometry estimates of similarly aged urban trees as documented in existing growth models?

  • 3) What is the potential for highway trees to provide carbon storage, carbon sequestration, stormwater reduction, and air pollutant removal according to ecosystem service estimation programs such as i-Tree (USDA; Madison, WI, USA)?

The results of this study can inform decisions about species and planting-site selection for highway planting projects in order to maximize the potential growth and benefits of these trees.

MATERIALS AND METHODS

The study trees were located next to Interstate 355 (I-355) in the greater Chicago region (United States; 41.932118, −88.037547; 41.538545, −87.960314) (Figure 1). This region has a temperate, continental climate with 10 °C average annual temperature; daily temperature extremes ranging from 40 °C to −32 °C; 94 cm average annual rainfall; and a median growing season of 177 days (Illinois State Climatologist 2007). The natural soils in the study area formed on glacial deposits, primarily moraines and till plains (Calsyn 1999; Hanson 2004). I-355 has 6 lanes, runs primarily north to south for 48 km, and has an average daily traffic volume of 271,980 cars per day (Illinois State Toll Highway Authority 2019).

Figure 1.
Figure 1.

Map of study area in Chicago metropolitan region of Illinois, USA, which shows the location of study trees from the 1988, 1997, and 2008 cohorts (Panel a). Also shown are example trees from the 1988 cohort (Panel b) and the 2008 cohort (Panel c).

To locate the I-355 trees, we reviewed landscaping drawings that were annotated to indicate the final quantity and species of planted trees (i.e., as-built landscape record drawings) from 3 planting projects on I-355 from 1988, 1997, and 2008. These plans documented a total of 14,806 total trees planted along the highway during these 3 time periods. We refer to the study trees as part of the 1988, 1997, and 2008 planting cohorts, which are based on the year when the landscaping projects were begun. The 1988 and 1997 cohorts are located in the northern portion of I-355 while the 2008 cohort is located in the southern, newer section of the highway. Northern I-355 is an older stretch of highway in a more densely developed area and typically has a narrower ROW with more sloped planting areas compared to the southern section. In a survey of the I-355 trees in 2018, we documented survival of trees from a randomly selected subset representing 20% of the total 14,806 trees documented in the landscape drawings for the 3 cohorts (Salisbury et al. 2022). Of this subset, 1,675 out of 2,944 surveyed trees (57%) were still alive in 2018.

The as-built records documented the location of planting sites along the highway and the quantity of trees originally in each planting site. Within a given planting site, all trees were the same species (or cultivar in some cases) and were the same stock size (recorded as nursery caliper, tree height, or container size). Planting sites typically contained between 5 to 50 trees which were planted in rows, usually no more than 3 trees deep with trees approximately 4.5 m to 6 m apart. For the purposes of analysis in this study, the planting site can be considered analogous to the concept of a stand in forest ecology. The planting sites randomly selected for this study resulted in the analysis of a total of 29 tree species (Table A1). Records of the time of year when the trees were planted as well as irrigation practices for the projects were not available.

In Spring 2019, we measured the trunk diameter at breast height (DBH) of 1,058 trees from randomly selected planting sites that had been used in the 2018 survival survey (Salisbury et al. 2022). Tree location was recorded using a GPS unit with sub-meter accuracy (Trimble TDC100 Handheld Unit with R1 GNSS Receiver; Westminster, CO, USA). We measured trees with single stems and codominant or multiple forked stems where pith separation occurred aboveground and below 1.4 m (multi-stemmed trees)(United States Forest Service 2021). In all cases, up to 6 of the largest stems with a diameter greater than 2.5 cm at 1.4 m above the ground surface were measured (United States Forest Service 2021). Multi-stem DBH was calculated using a quadratic sum equation (Equation 1) where DBHi is the DBH of an individual stem and DBHMS is the DBH-equivalent of the entire multistem tree (Magarik et al. 2020): Embedded Image 1

Dead stems were not measured and trees located on a slope were measured from the uphill side. Of the 1,058 DBH trees, 99% were planted as balled-and-burlap trees (excavated rootball wrapped in burlap, also known as caliper trees) while the remaining 1% were container grown (5-gallon [19-L] container or unspecified size). We report cohort age as the time elapsed since the year the planting project was started, recognizing that it is not possible to precisely know tree age in this setting because we lack records on the exact time each tree was planted and the trees’ ages at the time of planting.

GPS data was used to extract site-context data for each tree using ArcGIS (Version 10; ESRI; Redlands, CA, USA). An outline of the highway and associated ramps was created from aerial photos (United States Department of Agriculture 2017) while a digital elevation model (DEM) of the ROW with 1.2-m resolution was created using light detection and ranging (LiDAR) data (Illinois Geospatial Data Clearing-House 2015). To reduce the effects of potential errors in the combination of spatial data from different sources, we extracted average elevation, slope, and aspect data from a 2-m radius around each GPS point. For each tree, the geospatial data were used to determine the distance from the highway edge to the tree, the elevation of the tree relative to the highway, the slope of the ground, and the slope’s aspect (Figure 2). Aspect is the compass direction a slope faces as observed by a viewer facing downslope. Aspect is a circular variable and was decomposed into northing and easting components (the cosine and sine of aspect, respectively) for analysis. A northing value of 1 indicates a perfectly north facing slope while −1 is south facing. An easting value of 1 indicated a slope faces east while −1 faces west.

Figure 2.
Figure 2.

Site factors used to predict tree size on I-355.

The effects of distance from the highway, elevation, slope, northing, easting, and stem type (single or multi-stem) on DBH were tested for each cohort using a mixed-effects linear model (Table 1). Species and planting site were set as random intercepts since it is reasonable to assume different species will grow at different rates, and we observed variability in tree size among planting sites containing the same species. Models were visually checked for normality, homoscedasticity, and independence of residuals, which showed no adjustments were needed. Multicollinearity of predictor variables was checked using generalized variance inflation factors (GVIF)(Zuur et al. 2007). GVIF for all variables was less than 3, indicating multicollinearity was not an issue. Variable significance was assessed using a χ2 test that compared the full- and single-term deletion model. Significance was determined using a threshold of P = 0.05. Model performance was assessed using Akaike Information Criterion (AIC)(Petrov and Csaki 1973); conditional R-squared, which accounts for the contribution of fixed and random effects; and marginal R-squared, which accounts for only fixed effects (Nakagawa and Schielzeth 2013). The intraclass correlation coefficient (ICC) is another statistic that quantifies the proportion of variance explained by random factors and can be calculated as conditional ICC (including both fixed and random effects) and adjusted ICC (random effects only)(Nakagawa et al. 2017). All analyses were conducted using R (Version 3.6.2)(The R Foundation 2019) and the following packages: dplyr (Wickham et al. 2020), tidyr (Wickham 2020), ggplot (Wickham 2016), lme4 (Bates et al. 2015), and performance (Lüdecke et al. 2021).

View this table:
Table 1.

Site context model variables.

For 14 of the 29 observed species, we estimated the expected DBH of the trees using DBH-age allometric equations developed by McPherson et al. (2016). These equations are based on measurements of thousands of urban park and street trees from across the conterminous United States for the 20 most abundant species in a set of climate regions (McPherson et al. 2016). These trees were sampled from a range of land uses, including residential (single- and multi-family); commercial; industrial; and parkland (Table A2). For species with allometric equations from multiple climate regions, we used the equation from the climate region most similar to northern Illinois. Recognizing that the I-355 trees were likely planted over multiple years following the start of each planting project (i.e., trees were likely planted between 1988 and 1990 for the 1988 cohort), we estimated tree size for trees between the ages of 27 and 30 years for the 1988 cohort, 17 to 20 for 1997, and 7 to 10 for 2008.

We also used the I-355 DBH data to estimate the per-tree carbon sequestration, carbon storage, pollution removal, and runoff removal provided by all of the measured trees using i-Tree (Eco Version 6)(United States Forest Service 2020a). Since tree height, crown dimensions, and crown-condition data were not collected, we relied on i-Tree’s estimates of these parameters using allometric equations. We also used i-Tree’s default value of 13% missing crown, which should not seriously affect carbon sequestration estimates since no dead or dying trees were measured in our study (United States Forest Service 2020b).

RESULTS

Each planting cohort had a different combination of site context variables in the final model (Table 2). Northing (the north-south component of slope aspect) was significant in both the 1988 and 2008 cohort models, though interestingly north-facing slope trees were more likely to have smaller DBH in the 1988 cohort but were more likely to be larger in the 2008 cohort. For the 1997 cohort, both elevation relative to the road and slope had positive effects, i.e., trees at higher elevations and on steeper slopes had larger DBH. Stem type was also a significant predictor for the 2008 cohort with single-stem trees more likely to be smaller than multi-stemmed. For all 3 cohorts, marginal R-squared (representing the contributions of fixed effects only) tended to be substantially lower than conditional R-squared (contributions of fixed and random effects). In the 1997 and 2008 models, species random effects variance was greater compared to the planting site random effect. In the 1988 model, species and planting site random effects variances were similar. The adjusted intraclass correlation (ICC, evaluated with only random effects) was higher than the conditional ICC (fixed and random effects) for the 1988 and 1997 full and reduced models. There was no difference between the adjusted and conditional ICC for the 2008 full and reduced models.

View this table:
Table 2.

Full, final reduced (Red), and null model results for context variables in the 1988, 1997, and 2008 planting cohorts.

Of the 8 species in the 1988 planting cohort with allometric age-DBH equations in the Urban Tree Database (McPherson et al. 2016), only Quercus macrocarpa Michx. and Gleditsia triacanthos f. inermis ‘Imperial’ had DBH measurements which were similar to predicted values (Figure 3). DBH for the remaining species was lower than would be expected based on the allometric equations. In the 1997 cohort, Gymnocladus dioicus (L.) K. Koch DBH was similar to predicted values, though this species’ DBH was smaller than expected in the 1988 and 2008 cohorts (Figure 4). For the 2008 cohort, Q. macrocarpa and Pinus nigra ‘Arnold Sentinel’ were the only 2 species whose DBH was within the range of predicted values out of the 10 species with predicted DBH (Figure 5). Notably, DBH was generally more variable in the 1988 cohort compared to the 2008 cohort with the standard deviations for each species ranging from 0.6 to 10.6 for 1988 and from 0.4 to 4.8 for 2008. Multistem trees tended to have larger DBH (Equation 1) compared to single-stem trees within a given species, with the exceptions of Celtis occidentalis L. (1988), Populus × robusta (1988), and G. triacanthos ‘Skyline’ (1997).

Figure 3.
Figure 3.

DBH of tree species in the 1988 cohort. Species are grouped by stock size recorded in planting plans. Circles represent DBH for each individual tree. Gray bars indicate the estimated DBH at age 27 to 30 years for species included in the Urban Tree Database (McPherson et al. 2016). ‘Shade.’ = Shademaster cultivar. Full species names in Table A1.

Figure 4.
Figure 4.

DBH of tree species in the 1997 cohort. Species are grouped by stock size recorded in planting plans. Circles represent DBH for each individual tree. Gray bars indicate the estimated DBH at age 17 to 20 years for species included in the Urban Tree Database (McPherson et al. 2016). Full species names in Table A1.

Figure 5.
Figure 5.

DBH of tree species in the 2008 cohort. Species are grouped by stock size recorded in planting plans. Circles represent DBH for each individual tree. Gray bars indicate the estimated DBH at age 7 to 10 years for species included in the Urban Tree Database (McPherson et al. 2016). ‘Silver Q.’ = Silver Queen cultivar; ‘Shade.’ = Shademaster cultivar; “5 g” = 5-gallon container. Full species names in Table A1.

Median per-tree estimated carbon sequestration and estimated carbon storage were fairly similar between the 22- and 31-year-old cohorts, while median per-tree estimated pollution removal and estimated runoff removal were higher in the 30-year-old cohort (Figure 6). The ranges of per-tree estimated benefits overlap among all 3 planting cohorts reflecting differences in species composition as well as tree size. For example, the 2008 cohort Acer × freemanii with a mean DBH of 13 ± 3 cm had 1.7-times more estimated carbon storage than the small-statured Acer ginnala Maxim. in the 1988 cohort with a mean DBH of 11 ± 1 cm (Table A1). For estimated air-pollution removal, 2008 P. nigra and 1988 Pinus strobus L. had similar DBH (17 ± 2 cm and 16 ± 0.6 cm, respectively) but the 2008 P. nigra had 1.4-times greater air-pollution removal rates compared to the 1988 P. strobus.

Figure 6.
Figure 6.

Per-tree carbon sequestration (C Seq.), carbon storage (C Stor.), air-pollution removal (Poll. Rem.), and runoff removal (Run. Rem.) of I-355 study trees in the 3 planting cohorts.

DISCUSSION

Contrary to our initial hypotheses, no single planting location characteristic had a consistent effect on DBH. We expected trees growing closer to the highway edge to have smaller DBH since highway soil conditions are often the worst close to the roadway (Akbar et al. 2012). Soils adjacent to road pavement tend to receive highway runoff and are more likely to have been subject to intensive modification during construction activities (Trammell et al. 2011). For trees in this study, distance effects may have been unimportant since they were planted at least 9 m away from the highway (typical clear zone distance), avoiding the soil in close proximity to the pavement (with the exception of a few G. triacanthos planted in 1988, which were growing above a small retaining wall adjacent to an exit ramp). Differences in elevation relative to the highway may reflect differences in microtopography among planting sites since roadside microtopography can produce variable soil conditions within the ROW (Karim and Mallik 2008). Elevation relative to the roadside can also correlate with de-icing salt damage, with greater damage occurring below grade (Munck et al. 2010). The slope and aspect of hillsides can influence tree growth by affecting soil moisture, microclimate, and soil formation (Stage and Salas 2007). Increasing slope had a positive effect on DBH for the 1997 cohort, suggesting that in this case slope steepness may be a proxy for another variable such as soil conditions influencing tree size. Northing was an important model variable for both the 1988 and 2008 cohorts, though it had opposite effects on the 2 groups. Smaller DBH on north-facing slopes (northing = 1), seen in the 1988 cohort, could reflect changes in the height-DBH relationships in response to aspect and limited light (Long et al. 2020). Larger DBH on south-facing slopes, observed in the 2008 cohort, could be the result of cooler temperatures and less moisture loss from the soil which would reduce water stress (Stage and Salas 2007). The planting location characteristic models for I-355 demonstrate the difficulty in providing generalized recommendations for optimal tree planting locations in the ROW and suggest there may be other factors exerting a stronger influence on DBH in this setting.

Observations of trunk diameter of 10- to 30-year planted highway trees suggest highway ROWs in the Midwestern United States are a challenging environment for tree growth. Most of the species in this study with allometric equations from the Urban Tree Database (McPherson et al. 2016) had smaller DBH compared to model estimates for similarly aged trees found in other urban settings (e.g., residential, commercial, and parks). Granted, not all species could be evaluated in this manner since they lacked allometric equations in the Urban Tree Database. This gap highlights a need to continue collecting allometric data for multiple species in multiple urban settings to improve our expectations for long-term urban tree growth. Though by contrast, in Florida, Quercus virginiana Mill. and Taxodium distichum (L.) Rich. caliper growth in open lawns and park settings was lower compared to highway medians, parking lots, and other urban sites 38 months after planting (Koeser et al. 2014). Findings from Koeser et al. (2014) indicated that many factors in addition to land use, including planting practices, maintenance, and other site conditions, were necessary to predict tree growth in different urban settings.

Poor soil conditions following construction is likely an important driver of reduced highway tree growth. Research on soil restoration in highway settings indicated high bulk density and low organic-matter content following construction impeded tree growth (Somerville et al. 2018; McGrath et al. 2020). Trees growing near a highway in Finland also exhibited changes in leaf anatomy typically associated with drought stress, possibly caused by roadside microclimate and soil conditions (Nikula et al. 2011). Drought stress is highly relevant in northern Illinois where approximately every 2 out of 10 years experience well below average rainfall in the growing season (Calsyn 1999). Additionally, climate models predict this region could experience greater temperatures in the future which would worsen short-term droughts (Wuebbles et al. 2021). Such periodic drought conditions may worsen stress caused by post-construction soils and limit tree growth, though soil restoration prior to planting can improve tree growth in compacted soils while reducing the need for irrigation (McGrath et al. 2020).

Weather-related stress may have also played a role in limited growth for some of these tree species. The region has experienced several droughts during the planting period for the cohorts. These include extreme droughts (Palmer Drought Severity Index [PDSI] less than −4) from 1988 to 1989, early in the establishment period for the 1988 cohort, and in 2012 as well as several moderate to severe droughts (PDSI between −2 and −4) between 1989 and 2015 (Illinois State Climatologist 2021). There were also several years with severe to extreme wet spells from 2008 to 2011 (PDSI greater than 3)(Illinois State Climatologist 2021). Such unusually dry and unusually wet weather patterns occurring during the planting and establishment periods for the 1988 and 2008 cohorts, respectively, could have also contributed to the less-than-expected growth for some species. Considering predictions of more frequent droughts and intense rain events for Illinois (Wuebbles et al. 2021), it is reasonable to expect such weather-related stressors will continue to be relevant to urban tree growth in the future. Additionally, management practices can influence survival and growth (Vogt et al. 2015), potentially offsetting or exacerbating unfavorable weather. Unfortunately, we lack records documenting practices such as irrigation and the season when trees were planted and cannot draw conclusions about their possible effects.

Highway environments can contain elevated levels of de-icing materials in cold-weather climates (Fay and Shi 2012), heavy metals in roadside soil (Werkenthin et al. 2014), and other air-borne (Kuttler and Strassburger 1999) and soil pollutants (Bryselbout et al. 2000; Marusenko et al. 2011), which all have the potential to affect tree growth. Exposure to sodium-based de-icing salts has been associated with diminished stem growth in roadside trees, though these impacts vary among species and roadside conditions (Blomqvist 1998). Street trees (Tipuana tipu [Benth.] Kuntze) in São Paulo, Brazil, had reduced annual increment growth in association with elevated levels of particulate matter (PM10), zinc, barium, and aluminum (Locosselli et al. 2019). By contrast, in Finland, Populus tremula L. × P. tremuloides Michx. clones growing along highways had greater metal concentrations compared to rural sites, though concentrations were not at levels considered toxic and likely did not severely impact tree growth (Nikula et al. 2011). Considering the many potential stressors trees can encounter in the highway roadside in addition to other challenges associated with transplanting trees (Watson 2014), it is likely that multiple stressors could be impairing growth in the highway setting. Though notably, ameliorating compaction in highway soil using deep ripping or subsoiling and the incorporation of compost improved tree growth (Somerville et al. 2018; McGrath et al. 2020), offering a potential approach for alleviating at least some stressors in the roadside.

Stressful conditions in the highway ROW not only impair growth, they also limit trees’ ability to provide size-dependent ecosystem services. Nevertheless, highway trees have the potential to contribute substantial ecosystem services. For instance, the 303,000 trees in the United Kingdom’s Area 1, a 972-ha ROW (or soft estate), were estimated to annually remove 29 tons of air pollution, sequester 1,980 tons of carbon, and avoid over 75,000 m3 of runoff (Rogers and Evans 2015). And in Florida the 37,660 ha of vegetated ROW in the State Highway System (estimated to be about 40% woody vegetation) were estimated to provide approximately $5.95 million in air-pollution removal, $39.5 million in carbon sequestration, and $465 million in runoff reduction (USD, 2014 prices)(Harrison 2014). In our study, we observed that small-stature 31-year-old trees (e.g., Malus spp.) and large-stature 10-year-old trees (e.g., Acer saccharinum) had similar estimated carbon sequestration rates and storage quantities. This observation highlights how even after multiple decades, species with small-growth forms contribute little to carbon storage and sequestration, though small-statured species can provide other benefits such as aesthetics and can be appropriate for sites with space constraints. Additionally, size alone is not the only factor which influences ecosystem service provision. Q. macrocarpa on average is estimated to store and sequester the most carbon on a per-tree basis in the 1988 cohort, while other 1988 species have greater contributions to estimated avoided runoff. With the proliferation of large-scale tree planting initiatives across the globe (Eisenman et al. 2021), our results illustrate the role species selection can play in a planting project’s ability to provide meaningful ecosystem services. Indeed, in order for highway plantings to effectively provide particular benefits such as air-pollution removal, thoughtful species selection and arrangement are necessary (Baldauf 2017; Barwise and Kumar 2020) in addition to mitigating site stressors such as poor soil conditions.

Study Limitations

Diameter is only 1 metric of tree performance, and on its own does not capture long-term growth patterns or a tree’s overall condition (North et al. 2018). However, for the purposes of this research, diameter provided an efficient way to compare large numbers of similarly aged trees. Planting density is 1 factor which influences a tree’s allocation of resources to growing taller or wider, consequently influencing caliper (Drew and Flewelling 1979). For the I-355 planting sites, tree spacing was consistent among planting sites (ranging from 4.6 m to 6 m) and tree planting areas were rarely more than 2 rows wide, suggesting edge and interior effects would be negligible. While there are multiple methods for measuring the DBH of multi-stemmed trees with different advantages and disadvantages (Magarik et al. 2020), we used the 6-largest-stems approach in order to be consistent with i-Tree protocol (United States Forest Service 2021). Since cohort age is based on the year the planting projects started and we lack records of exactly when each tree was planted, cohort age may overestimate the amount of time a tree has been in its planting site. The accuracy of i-Tree estimates of ecosystem services are limited by the applicability of their allometric models to particular local conditions (Timilsina et al. 2017). Consequently, comparing estimates as ratios or percentages rather than absolute values is likely a more informative approach.

CONCLUSION

The highway ROW presents many challenges for planting and growing trees, but also great potential. We evaluated the current size of 3 planting cohorts along a highway in the Chicago metropolitan region to assess factors which could influence long-term tree growth. Our observations demonstrated that planting location within the ROW can impact tree size and consequently some of the ecosystem services provided by those trees. However, the effects of planting location on tree size were inconsistent among tree species, and it is difficult to draw generalized recommendations about choosing planting sites within the ROW from the data. These results emphasize the point that highway ROWs are not a homogenous planting environment. Consequently, careful on-the-ground site assessments and soil restoration when needed should increase the likelihood of trees realizing their full potential for growth along highways.

ACKNOWLEDGMENTS

This project was made possible with support from the Illinois State Toll Highway Authority. We thank Andrew Muñoz-Male for assistance with data collection, Tana Perez for her data anaylsis work, as well as the rest of The Morton Arboretum’s Center for Tree Science team for their support. CREDIT Roles: ABS: conceptualization, data curation, formal analysis, investigation, methodology, visualization, writing – original draft, writing – review and editing. JWM: conceptualization, methodology, supervision, project administration, funding acquisition, writing – review and editing. AKK: formal analysis, writing – review and editing.

Appendix 1.

View this table:
Table A1.

Mean and standard deviation (in parentheses) of DBH and ecosystem services for each species and year. C seq. = carbon sequestration. C stor. = carbon storage. Poll. rem. = air-pollution removal. Run. rem. = runoff removal.

View this table:
Table A2.

Formulas used to estimate DBH based on tree age using data from the Urban Tree Database for particular species obsen/ed in different regions of the United States (McPherson et al. 2016).

Footnotes

  • Conflicts of Interest:

    The authors reported no conflicts of interest.

LITERATURE CITED

  1. ↵
    1. Akbar KF,
    2. Hale WHG,
    3. Šerá B,
    4. Ashraf I
    . 2012. Phytometric assessment of fertility of roadside soils and its relationship with major nutrients. Polish Journal of Environmental Studies. 21(5):1141–1145. http://www.pjoes.com/pdf-88850-22709?filename=22709.pdf
    OpenUrl
  2. ↵
    1. Baldauf R
    . 2017. Roadside vegetation design characteristics that can improve local, near-road air quality. Transportation Research Part D: Transport and Environment. 52(Part A):354–361. https://doi.org/10.1016/j.trd.2017.03.013
    OpenUrl
  3. ↵
    1. Barwise Y,
    2. Kumar P
    . 2020. Designing vegetation barriers for urban air pollution abatement: A practical review for appropriate plant species selection. npj Climate and Atmospheric Science. 3:1–19. https://doi.org/10.1038/s41612-020-0115-3
    OpenUrl
  4. ↵
    1. Bates D,
    2. Mächler M,
    3. Bolker B,
    4. Walker S
    . 2015. Fitting linear mixed-effects models using lme4. Journal of Statistical Software. 67(1):1–48. https://doi.org/10.18637/jss.v067.i01
    OpenUrlCrossRefPubMed
  5. ↵
    1. Blomqvist G
    . 1998. Impact of de-icing salt on roadside vegetation: A literature review. VTI Rapport. Stockholm (Sweden): Swedish National Road and Transport Research Institute. 35 p. http://www.diva-portal.org/smash/get/diva2:675169/FULLTEXT01.pdf
  6. ↵
    1. Bryselbout C,
    2. Henner P,
    3. Carsignol J,
    4. Lichtfouse É
    . 2000. Polycyclic aromatic hydrocarbons in highway plants and soils. Evidence for a local distillation effect. Analusis. 28(4):290–293. https://doi.org/10.1051/analusis:2000280290
    OpenUrl
  7. ↵
    1. Bryson GM,
    2. Barker AV
    . 2002. Sodium accumulation in soils and plants along Massachusetts roadsides. Communications in Soil Science and Plant Analysis. 33(1-2):67–78. https://doi.org/10.1081/CSS-120002378
    OpenUrl
  8. ↵
    1. Calsyn DE
    . 1999. Soil survey of Du Page County, Illinois. Washington (DC, USA): United States Department of Agriculture, Natural Resources Conservation Service. 253 p. https://www.nrcs.usda.gov/Internet/FSE_MANUSCRIPTS/illinois/IL043/0/Du_Page_IL.pdf
  9. ↵
    1. Chen Y,
    2. Wang X,
    3. Jiang B,
    4. Wen Z,
    5. Yang N,
    6. Li L
    . 2017. Tree survival and growth are impacted by increased surface temperature on paved land. Landscape and Urban Planning. 162:68–79. https://doi.org/10.1016/j.landurbplan.2017.02.001
    OpenUrl
  10. ↵
    1. De Lacy P,
    2. Shackleton CM
    . 2014. The comparative growth rates of indigenous street and garden trees in Grahamstown, South Africa. South African Journal of Botany. 92:94–96. https://doi.org/10.1016/j.sajb.2014.02.007
    OpenUrl
  11. ↵
    1. Drew TJ,
    2. Flewelling JW
    . 1979. Stand density management: An alternative approach and its application to Douglas-fir plantations. Forest Science. 25(3):518–532. https://doi.org/10.1093/forestscience/25.3.518
    OpenUrl
  12. ↵
    1. Eisenman TS,
    2. Flanders T,
    3. Harper RW,
    4. Hauer RJ,
    5. Lieberknecht K
    . 2021. Traits of a bloom: A nationwide survey of U.S. urban tree planting initiatives (TPIs). Urban Forestry & Urban Greening. 61:127006. https://doi.org/10.1016/j.ufug.2021.127006
    OpenUrl
  13. ↵
    1. Fay L,
    2. Shi X
    . 2012. Environmental impacts of chemicals for snow and ice control: State of the knowledge. Water, Air, & Soil Pollution. 223:2751–2770. https://doi.org/10.1007/s11270-011-1064-6
    OpenUrl
  14. ↵
    1. Forman RTT,
    2. Sperling D,
    3. Bissonette JA,
    4. Clevenger AP,
    5. Cutshall CD,
    6. Dale VH,
    7. Fahrig L,
    8. France RL,
    9. Goldman CR,
    10. Heanue K,
    11. Jones J,
    12. Swanson F,
    13. Turrentine T,
    14. Winter TC
    . 2002. Road ecology: Science and solutions. Illustrated Edition. Washington (DC, USA): Island Press. 504 p.
  15. ↵
    1. Hanson KD
    . 2004. Soil survey of Will County, Illinois. Washington (DC, USA): United States Department of Agriculture, Natural Resources Conservation Service. 544 p. https://www.nrcs.usda.gov/Internet/FSE_MANUSCRIPTS/illinois/IL197/0/will_IL.pdf
  16. ↵
    1. Harrison GL
    . 2014. Economic impact of ecosystem services provided by ecologically sustainable roadside right of way vegetation management practices. Tallahassee (FL, USA): Florida Department of Transportation. FDOT Contract No: BDK75-977-74. 35 p. https://rightofway.erc.uic.edu/wp-content/uploads/2018/05/2A4-FDOT-ecosystem-services-roadsides-report.pdf
  17. ↵
    1. Hilbert DR,
    2. Roman LA,
    3. Koeser AK,
    4. Vogt J,
    5. van Doorn NS
    . 2019. Urban tree mortality: A literature review. Arboriculture & Urban Forestry. 45(5):167–200. https://doi.org/10.48044/jauf.2019.015
    OpenUrl
  18. ↵
    1. Hodge SJ,
    2. Boswell R
    . 1993. A study of the relationship between site conditions and urban tree growth. Journal of Arboriculture. 19(6):358–367. https://doi.org/10.48044/jauf.1993.056
    OpenUrl
  19. ↵
    1. Illinois Geospatial Data ClearingHouse
    . 2015. Illinois height modernization (ILHMP): LiDAR data. Champaign (IL, USA): Illinois State Geological Survey. [Accessed 2019 October 12]. https://clearinghouse.isgs.illinois.edu/data/elevation/illinois-height-modernization-ilhmp
  20. ↵
    1. Illinois State Climatologist
    . 2007. Joliet Station 114530: Precipitation summary. Champaign (IL, USA): Illinois State Climatologist. [Updated 2007 September 11; Accessed 2020 March 31]. https://stateclimatologist.web.illinois.edu/data/illinois-climate-summaries/joliet-station-114530
  21. ↵
    1. Illinois State Climatologist
    . 2021. Drought trends. Champaign (IL, USA): Illinois State Climatologist. [Accessed 2022 February 1]. https://stateclimatologist.web.illinois.edu/drought-in-illinois/drought-trends
  22. ↵
    1. Illinois State Toll Highway Authority
    . 2019. 2019 traffic data report for the Illinois Tollway System. Boston (MA, USA): CDM Smith. 251 p. https://www.illinoisvirtualtollway.com/Traffic/2019Traffic_Data_Report.pdf
  23. ↵
    1. Jimenez MD,
    2. Ruiz-Capillas P,
    3. Mola I,
    4. Pérez-Corona E,
    5. Casado MA,
    6. Balaguer L
    . 2011. Soil development at the roadside: A case study of a novel ecosystem. Land Degradation & Development. 24(6):564–574. https://doi.org/10.1002/ldr.1157
    OpenUrl
  24. ↵
    1. Karim MN,
    2. Mallik AU
    . 2008. Roadside revegetation by native plants: I. Roadside microhabitats, floristic zonation and species traits. Ecological Engineering. 32(3):222–237. https://doi.org/10.1016/j.ecoleng.2007.11.003
    OpenUrl
  25. ↵
    1. Koeser AK,
    2. Gilman EF,
    3. Paz M,
    4. Harchick C
    . 2014. Factors influencing urban tree planting program growth and survival in Florida, United States. Urban Forestry & Urban Greening. 13(4):655–661. https://doi.org/10.1016/j.ufug.2014.06.005
    OpenUrl
  26. ↵
    1. Kuttler W,
    2. Strassburger A
    . 1999. Air quality measurements in urban green areas—A case study. Atmospheric Environment. 33(24-25):4101–4108. https://doi.org/10.1016/S1352-2310(99)00151-X
    OpenUrlCrossRef
  27. ↵
    1. Lawrence AB,
    2. Escobedo FJ,
    3. Staudhammer CL,
    4. Zipperer W
    . 2012. Analyzing growth and mortality in a subtropical urban forest ecosystem. Landscape and Urban Planning. 104(1):85–94. https://doi.org/10.1016/j.landurbplan.2011.10.004
    OpenUrlCrossRef
  28. ↵
    1. Leff M
    . 2016. The sustainable urban forest: A step-by-step approach. Washington (DC, USA): United States Department of Agriculture Forest Service. 109 p.
  29. ↵
    1. Locosselli GM,
    2. de Camargo EP,
    3. Moreira TCL,
    4. Todesco E,
    5. de Fátima Andrade M,
    6. de André CDS,
    7. de André PA,
    8. Singer JM,
    9. Ferreira LS,
    10. Saldiva PHN,
    11. Buckeridge MS
    . 2019. The role of air pollution and climate on the growth of urban trees. Science of the Total Environment. 666:652–661. https://doi.org/10.1016/j.scitotenv.2019.02.291
    OpenUrl
  30. ↵
    1. Long S,
    2. Zeng S,
    3. Liu F,
    4. Wang G
    . 2020. Influence of slope, aspect and competition index on the height-diameter relationship of Cyclobalanopsis glauca trees for improving prediction of height in mixed forests. Silva Fennica. 54(1):1–20. https://doi.org/10.14214/sf.10242
    OpenUrl
  31. ↵
    1. Lüdecke D,
    2. Ben-Shachar MS,
    3. Patil I,
    4. Waggoner P,
    5. Makowski D
    . 2021. Performance: An R package for assessment, comparison and testing of statistical models. Journal of Open Source Software. 6(60):3139. https://doi.org/10.21105/joss.03139
    OpenUrl
  32. ↵
    1. Magarik YAS,
    2. Roman LA,
    3. Henning JG
    . 2020. How should we measure the DBH of multi-stemmed urban trees? Urban Forestry & Urban Greening. 47:126481. https://doi.org/10.1016/j.ufug.2019.126481
    OpenUrl
  33. ↵
    1. Marusenko Y,
    2. Herckes P,
    3. Hall SJ
    . 2011. Distribution of polycyclic aromatic hydrocarbons in soils of an arid urban ecosystem. Water, Air, & Soil Pollution. 219:473–487. https://doi.org/10.1007/s11270-010-0721-5
    OpenUrl
  34. ↵
    1. McGrath D,
    2. Henry J,
    3. Munroe R,
    4. Williams C
    . 2020. Compost improves soil properties and tree establishment along highway roadsides. Urban Forestry & Urban Greening. 55:126851. https://doi.org/10.1016/j.ufug.2020.126851
    OpenUrl
  35. ↵
    1. McPherson EG,
    2. van Doorn NS,
    3. Peper PJ
    . 2016. Urban tree database. Fort Collins (CO, USA): Forest Service Research Data Archive. [Updated 2020 January 21]. https://doi.org/10.2737/RDS-2016-0005
  36. ↵
    1. Munck IA,
    2. Bennett CM,
    3. Camilli KS,
    4. Nowak RS
    . 2010. Long-term impact of de-icing salts on tree health in the Lake Tahoe Basin: Environmental influences and interactions with insects and diseases. Forest Ecology and Management. 260(7):1218–1229. https://doi.org/10.1016/j.foreco.2010.07.015
    OpenUrlCrossRef
  37. ↵
    1. Nakagawa S,
    2. Johnson PCD,
    3. Schielzeth H
    . 2017. The coefficient of determination R2 and intra-class correlation coefficient from generalized linear mixed-effects models revisited and expanded. Journal of the Royal Society Interface. 14:20170213. https://doi.org/10.1098/rsif.2017.0213
    OpenUrl
  38. ↵
    1. Nakagawa S,
    2. Schielzeth H
    . 2013. A general and simple method for obtaining R2 from generalized linear mixed-effects models. Methods in Ecology and Evolution. 4(2):133–142. https://doi.org/10.1111/j.2041-210x.2012.00261.x
    OpenUrl
  39. ↵
    1. Neher DA,
    2. Asmussen D,
    3. Lovell ST
    . 2013. Roads in northern hardwood forests affect adjacent plant communities and soil chemistry in proportion to the maintained roadside area. Science of the Total Environment. 449:320–327. https://doi.org/10.1016/j.scitotenv.2013.01.062
    OpenUrl
  40. ↵
    1. Nikula S,
    2. Manninen S,
    3. Vapaavuori E,
    4. Pulkkinen P
    . 2011. Growth, leaf traits and litter decomposition of roadside hybrid aspen (Populus tremula L. × P. tremuloides Michx.) clones. Environmental Pollution. 159(7):1823–1830. https://doi.org/10.1016/j.envpol.2011.03.022
    OpenUrlPubMed
  41. ↵
    1. North EA,
    2. D’Amato AW,
    3. Russell MB
    . 2018. Performance metrics for street and park trees in urban forests. Journal of Forestry. 116(6):547–554. https://doi.org/10.1093/jofore/fvy049
    OpenUrl
  42. ↵
    1. Nowak DJ,
    2. Crane DE,
    3. Stevens JC,
    4. Hoehn RE,
    5. Walton JT,
    6. Bond J
    . 2008. A ground-based method of assessing urban forest structure and ecosystem services. Arboriculture & Urban Forestry. 34(6):347–358. https://doi.org/10.48044/jauf.2008.048
    OpenUrl
  43. ↵
    1. Petrov BN,
    2. Csaki F
    , editors. 1973. Second international symposium on information theory. 1971 September 2–8; Tsahkadsor, Armenia, USSR. Budapest (Hungary): Akademiai Kiado. 451 p.
  44. ↵
    1. The R Foundation
    . 2019. The R Project for Statistical Computing. Vienna (Austria): The R Foundation. https://www.R-project.org
  45. ↵
    1. Rhoades RW,
    2. Stipes RJ
    . 1999. Growth of trees on the Virginia Tech campus in response to various factors. Journal of Arboriculture. 25(4):211–217. https://doi.org/10.48044/jauf.1999.031
    OpenUrl
  46. ↵
    1. Rogers K,
    2. Evans G
    . 2015. Valuing the natural capital of Area 1: A pilot study. Birmingham (UK): Highways Agency. 54 p.
  47. ↵
    1. Roman LA,
    2. Conway TM,
    3. Eisenman TS,
    4. Koeser AK,
    5. Ordóñez Barona C,
    6. Locke DH,
    7. Jenerette GD,
    8. Östberg J,
    9. Vogt J
    . 2021. Beyond ‘trees are good’: Disservices, management costs, and tradeoffs in urban forestry. Ambio. 50:615–630. https://doi.org/10.1007/s13280-020-01396-8
    OpenUrl
  48. ↵
    1. Salisbury AB,
    2. Miesbauer JW,
    3. Koeser AK
    . 2022. Long-term tree survival and diversity of highway tree planting projects. Urban Forestry & Urban Greening. 73:12754. https://doi.org/10.1016/j.ufug.2022.127574
    OpenUrl
  49. ↵
    1. Smith IA,
    2. Dearborn VK,
    3. Hutyra LR
    . 2019. Live fast, die young: Accelerated growth, mortality, and turnover in street trees. PLoS ONE. 14(5):e0215846. https://doi.org/10.1371/journal.pone.0215846
    OpenUrlCrossRef
  50. ↵
    1. Somerville PD,
    2. May PB,
    3. Livesley SJ
    . 2018. Effects of deep tillage and municipal green waste compost amendments on soil properties and tree growth in compacted urban soils. Journal of Environmental Management. 227:365–374. https://doi.org/10.1016/j.jenvman.2018.09.004
    OpenUrl
  51. ↵
    1. Stage AR,
    2. Salas C
    . 2007. Interactions of elevation, aspect, and slope in models of forest species composition and productivity. Forest Science. 53(4):486–492. https://doi.org/10.1093/forestscience/53.4.486
    OpenUrl
  52. ↵
    1. Steenberg JWN,
    2. Millward AA,
    3. Nowak DJ,
    4. Robinson PJ
    . 2017. A conceptual framework of urban forest ecosystem vulnerability. Environmental Reviews. 25(1):115–126. https://doi.org/10.1139/er-2016-0022
    OpenUrl
  53. ↵
    1. Timilsina N,
    2. Beck JL,
    3. Eames MS,
    4. Hauer R,
    5. Werner L
    . 2017. A comparison of local and general models of leaf area and biomass of urban trees in USA. Urban Forestry & Urban Greening. 24:157–163. https://doi.org/10.1016/j.ufug.2017.04.003
    OpenUrl
  54. ↵
    1. Trammell TLE,
    2. Schneid BP,
    3. Carreiro MM
    . 2011. Forest soils adjacent to urban interstates: Soil physical and chemical properties, heavy metals, disturbance legacies, and relationships with woody vegetation. Urban Ecosystems. 14:525–552. https://doi.org/10.1007/s11252-011-0194-3
    OpenUrl
  55. ↵
    1. Turner-Skoff JB,
    2. Cavender N
    . 2019. The benefits of trees for livable and sustainable communities. Plants, People, Planet. 1(4):323–335. https://doi.org/10.1002/ppp3.39
    OpenUrl
  56. ↵
    1. United States Department of Agriculture
    . 2017. National agriculture imagery program (NAIP). Information Sheet. Washington (DC, USA): United States Department of Agriculture. 2 p. [Accessed 2019 October 10]. https://www.fsa.usda.gov/Assets/USDA-FSA-Public/usdafiles/APFO/support-documents/pdfs/naip_infosheet_2017.pdf
  57. ↵
    1. United States Forest Service
    . 2020a. i-Tree Eco v6. [Accessed 2020 July 8]. https://www.itreetools.org/tools/i-tree-eco
  58. ↵
    1. United States Forest Service
    . 2020b. Eco guide to data limitations. 3 p. https://www.itreetools.org/documents/263/DataLimitations.2020_07_15.pdf
  59. ↵
    1. United States Forest Service
    . 2021. i-Tree Eco field guide. v6.0.57 p. [Updated 2021 October 6]. https://www.itreetools.org/documents/274/EcoV6.FieldManual.2021.10.06.pdf
  60. ↵
    1. Vogt JM,
    2. Watkins SL,
    3. Mincey SK,
    4. Patterson MS,
    5. Fischer BC
    . 2015. Explaining planted-tree survival and growth in urban neighborhoods: A social-ecological approach to studying recently-planted trees in Indianapolis. Landscape and Urban Planning. 136:130–143. https://doi.org/10.1016/j.landurbplan.2014.11.021
    OpenUrl
  61. ↵
    1. Watson G
    . 2014. Best management practices—Tree planting. 2nd Ed. Champaign (IL, USA): International Society of Arboriculture. 40 p.
  62. ↵
    1. Werkenthin M,
    2. Kluge B,
    3. Wessolek G
    . 2014. Metals in European roadside soils and soil solution—A review. Environmental Pollution. 189:98–110. https://doi.org/10.1016/j.envpol.2014.02.025
    OpenUrlCrossRef
  63. ↵
    1. Wickham H.
    2016. ggplot2: Elegant graphics for data analysis. 2nd Ed. Cham (Switzerland): Springer International Publishing AG. 276 p.
  64. ↵
    1. Wickham H.
    2020. tidyr: Tidy messy data. https://tidyr.tidyverse.org
    1. Wickham H,
    2. François R,
    3. Henry L,
    4. Müller K
    . 2020. dplyr: A grammar of data manipulation. https://dplyr.tidyverse.org
  66. ↵
    1. Wuebbles D,
    2. Angel J,
    3. Petersen K,
    4. Lemke AM
    . 2021. 2021: An assessment of the impacts of climate change in Illinois. Champaign (IL, USA): University of Illinois at Urbana-Champaign. https://doi.org/10.13012/B2IDB-1260194_V1
  67. ↵
    1. Zuur AF,
    2. Ieno EN,
    3. Smith GM
    . 2007. Analyzing ecological data. New York (NY, USA): Springer. 698 p.
Back to top

In this issue

Print
Download PDF
Email Article
Citation Tools
Long-Term Growth of Highway Rights-of-Way Trees
Allyson B. Salisbury, Jason W. Miesbauer, Andrew K. Koeser
Arboriculture & Urban Forestry (AUF) Sep 2022, 48 (5) 293-307; DOI: 10.48044/jauf.2022.022
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo
Bookmark this article

Jump to section

Keywords