Newly Planted Street Tree Growth and Mortality

Methods

Four hundred eighty street trees were planted in the fall of 1985 along 5.4 miles of Martin Luther King Blvd, extending from southern Berkeley through western Oakland, California to San Francisco Bay. Volunteers planted 254 black locusts (Robinia pseudoacacia ‘Purple Robe’); 199 southern magnolias (Magnolia grandiflora ‘Samuel Sommer’) and 27 London plane trees (Platanus x acerífolía). The majority of these trees were double staked after planting, although some center nursery stakes were left attached.

In January of 1986, 1987 and 1988, all trees were measured with regard to: 1) diameter at 4.5 feet above the curb (to the nearest 1/32 inch); 2) tree height; 3) height to lowest branch; 4) branch length parallel and perpendicular to the street; 5) vandalism and 6) mortality. Vandalized trees were defined as trees in which the main stem had broken or the main stems were cut off above the tree base. This classification includes both intentionally and accidentally damaged trees, though most trees were thought to be intentionally damaged. Relatively minor vandalism acts (e.g., broken branches) were not included in this classification.

Additional information was also collected on: 1) planting strip size; 2) height of and distance to trees within two competitor’s tree heights; and 3) adjacent land use. Planting strip size was calculated by multiplying the planting width not covered by impervious surfaces by planting length. For continuous planting strips, a length of twice the average tree height (2 × 2.97 m) was substituted.

The boulevard was stratified by census tracts (3) and socio-economic data were collected for each of 9 different census tract areas. The socioeconomic data included percent of population of black persons; percent of population of white persons; median house value; median monthly rent; population density; percent of population less than 25 years old; percent unemployment; per capita income; percent of population with high school degrees; and ratio of owner occupied housing units to rented housing units. Areas where Martin Luther King Blvd. was a border between two census tracts, weighted averages of the two tracts were used to obtain socioeconomic data.

A root competition index was devised to assess the relative impact of competing trees. Any two trees were considered competing if the distance between them was less than two times the tallest tree’s height. An index value was calculated for every tree competing with a street tree as: [2 / (distance to tree / competitor’s height)] - 1. For each street tree, these values were summed for all competing trees to yield a relative root competition index.

Crown volume was estimated for each tree using standard geometric formulas and measurements of crown height (tree height -height to lowest branch) and average branch length (crown radius). Black locust’s crown was characterized by a cylinder (V = (pi) (r²) (h)); magnolia as an ellipsoid (V = (4/3) (pi) (O.5h) (road parallel radius) (road perpendicular radius)); and London plane tree as a paraboloidal segment (V = (1/2) (pi) (r≡) (h)).

City maintenance practices were generally limited to pruning, restaking and replanting (B. Jones and J. Ryugo, personal communication, 1988). Watering and fertilization (if any) were left up to adjacent homeowners. In Oakland, trees in the predominantly commercial/industrial areas (i.e., few homeowners) were watered by city personnel during summer months (J. Ryugo, personal communication, 1988).

Tests of significance employed were paired t-tests to test for species growth differences between years, 1-way analysis of variance with Scheffe test to separate the growth and diameter means among species, t-tests for signficance of Pearson’s r and chi-square tests for signficant differences in mortality and vandalism. The level for statistical significance was set at alpha = 0.05.

Results

Discussion

Growth and diameter differed significantly among the species with black locust and London plane tree exhibiting larger growth and diameters than magnolias. Although there was no difference in diameters of the dead trees in comparison with the live trees, this lack of difference is likely because all the trees are relatively small. As diameters increase beyond a certain caliper, mortality may decrease because these larger trees are better able to withstand vandalism and automobile damage. Planting relatively large street trees has been noted to help reduce vandalism (2). Thus, black locust and London plane tree mortality rates may drop sooner than magnolia’s mortality rate.

To attain large trees more quickly, thereby potentially reducing young tree mortality, growth rates should be enhanced. Along with irrigation, fertilization and reducing competition, one way to enhance growth is to preserve crown volume (photosynthetic area) by minimizing pruning. Although pruning is an important part of early tree maintenance, excessive pruning should be avoided to maximize early total growth rates.

Staking techniques have also been noted to be a cause of newly planted tree mortality (5). The double staking technique (with a high crossbar) employed on Martin Luther King Blvd. may lead to increased mortality. With double staking, a high crossbar or a rigid pair of ties between the two stakes can act as a fulcrum at which trees are readily snapped by vandals. Single staking with a steel-reinforcing-bar at least 5 feet tall, tied close to the tree, is reported to reduce this problem (2), but may increase damage to the trunk due to the bark rubbing against the stake. Smooth steel stakes may reduce this type of damage.

Other staking techniques that may reduce vandalism are: 1) double staking without a high crossbar and one flexible tie to hold the tree upright, provide flexibility, and minimize tree injury; 2) triple staking; or 3) using sturdy stakes encircled with heavy wire or metal grillwork (6).

Our finding on first year mortality rates (18%) did not differ much from the average first year mortality found in the literature (approx. 16%) (4, 7). Our 2-year mortality findings (34%) were higher than what was found in Boston (23%) and were within the 3-year mortality range found in Oakland (30-40%) (4, 11).

Black (2) reported no link between Seattle’s vandalism and economic status or racial make-up of neighborhoods, but did note that most vandalized trees were broken by males age 17 to 25. Although males age 17 to 25 may cause the most vandalism, we found no association between the percent of the population less than 25 years old and mortality. Mortality also was not associated with race.

Mortality was negatively associated with socioeconomic status. Our sample transect though, did not include the more affluent areas of Berkeley or Oakland. The best socio-economic indicator of mortality appears to be percent unemployed. The areas of higher unemployment also generally had lower rents, lower per capita income, fewer owner-occupied homes, and fewer high school graduates.

Various reasons could be hypothesized as to why lower socio-economic areas had greater tree mortality. Increased unemployment signifies more time spent in the neighborhood and increased activity in the street tree environment, both social (pedestrian) and possibly automobile activity. Lower income indicates less funds available for tree care. Fewer owner-occupied residences connotes a lack of sense of ownership toward the street trees and therefore a possible lack of care. Fewer high school graduates implies that an increase in education may decrease mortality.

Berkeley’s lower mortality rates are likely due to differences in soci-economic status between the cities (the Berkeley sample area has a relatively higher socio-economic standing than most of the Oakland sample area). This difference may lead to greater homeowner care for Berkeley street trees.

Sklar and Ames (11) concluded the success of an Oakland inner-city tree planting program (which included neighborhood meetings, corporate sponsorship and ceremonial plantings) was not due to the program educating the residents. Rather, the tree planting program tended to define the parkway trees as part of the resident’s property, thus decreasing the significance of the parkwayproperty barrier. Future studies are needed to determine what factors or combination of factors are responsible for increased mortality.

Areas of lower socio-economic status also had significantly smaller planting spaces. Although differences in planting space size did not affect mortality, smaller planting spaces indicate a harsher physical tree environment.

Different adjacent land uses affected mortality in different ways. Apartments and public greenspaces were associated with increased mortality while single family residences and subway stations (BART) were associated with decreased mortality. The major differences between these land uses are likely the sense of ownership and care toward the trees and the amount of people concentrated about these trees..

In areas with apartments and public greenspaces, one would expect a lower sense of tree ownership and relatively high concentrations of people. The opposite would be expected for single family homes and BART stations. The street trees adjacent to BART were not near the station entrance and likely had low pedestrian flow. These trees were also likely maintained by BART personnel.

Conclusion

The social environment around the tree is just as important as the physical tree environment for insuring early tree survival. Management and maintenance activities of the urban forester can greatly impact early tree health and survival. Areas of lower socio-economic status, as well as public greenspaces and apartment complexes, should be identified for their increased potential for high mortality. In these areas, a more intensive effort of public education and involvement should be sustained. These efforts will likely promote a sense of ownership and pride toward the trees and decrease mortality. Planting larger trees and proper staking techniques may also aid in reducing mortality rates in problem areas.