Download exploring boston`s urban forest

2014-15
FIELD
REPORT
Amazon Riverboat Exploration—2012 FIELD REPORT
EXPLORING BOSTON’S
URBAN FOREST
1
Exploring Boston’s Urban Forest
2014/2015 FIELD REPORT
Background Information
LEAD PI: Dr. Vanessa Boukili
REPORT COMPLETED BY: Dr. Vanessa Boukili
DATE REPORT COMPLETED: 2016-04-01
DISCLAIMER/AUTHORS’ NOTE
Please do not use any of the data or results presented here without explicit permission from Dr. Vanessa Boukili. Many of the
results shown here are preliminary, and will not be final until publication. Most of these results will be included in the Urban
Forest Management Plan for the City of Cambridge (in preparation), or in journal articles. In the meantime, the results should
be interpreted with caution.
Phone: 1-800-776-0188
Web: earthwatch.org
facebook.com/earthwatch
twitter.com/earthwatch_org
Dear Earthwatch Volunteers,
Thank you so much for joining us on our expedition “Exploring Boston’s Urban Forest” in 2014
and 2015. I became the lead scientist on this project in early 2014, and I could not have developed
this program without you. I am incredibly grateful for the energy, enthusiasm, and excitement
you brought to the expedition, working hard in the field and quickly becoming adept at species
identification and diameter at breast height (DBH) measurements. You asked insightful questions,
and showed great interest in and concern for the trees in your backyard. Many of you came away
from the day with a new perspective—I often heard you saying you would never look at the trees
in the city the same way. You have changed my perspective too, and now when I look at the trees,
I see you measuring them and assessing their health. While walking or driving around Cambridge,
Somerville, Chelsea, or Boston I can actually point to specific trees that you measured, and it
brings a smile to my face.
But you brought so much more than enthusiasm to this expedition. We had 220 volunteers
in 2014, and 195 volunteers in 2015. With all of your hard work, we collected data on more than
4,000 trees in Cambridge, more than 1,000 trees in Somerville, and hundreds of trees each in
Chelsea and Boston. The data you collected was very accurate. Remember those first three
trees you and your group measured after the introduction? Across all 296 DBH measurements
that your group and other groups made on the quality control trees across the two years, only
5% of measurements were large errors (i.e., errors that differed from my measurements by more
than 10%). Of the remaining quality control measurements, the average difference between your
measurements and my measurements was very small, only 1.2%. And, since these measurements
were the first ones you made, it is likely that your measurements became even more accurate over
time! The data you collected is a great resource that does not exist anywhere else, and which will
continue to be used for years to come.
Unfortunately, in late 2015, Earthwatch made the difficult decision to retire this expedition due
to lack of funding. But, thanks to you, we consider this expedition a complete success. To date, the
data you collected have led to the creation of two scientific manuscripts. These data have also led
to a partnership with the City of Cambridge to create an Urban Forest Management Plan for the
city. These data are also being incorporated into the tree inventories for each city, which will help
the city managers continue to track these trees in the future.
Again, I thank each of you for the time and effort you spent with me on this expedition. I hope
that you will continue to appreciate the trees in your backyard, and remember all they do for the
environment, and for us. A city can be a very hard place for a tree to live, but we can help give
them their best chance.
Sincerely,
Dr. Vanessa Boukili
SUMMARY
Quality control analyses comparing the DBH measurements made by citizen scientists to measurements made by an expert
confirm that the citizen scientist data is largely accurate. For the 2014–2015 quality control analyses, citizen scientists made
296 measurements on 44 different trees that were also measured by an expert. Of 296 DBH measurements, only 15 (5%) were
large errors, differing from the expert measurement by 10% or more. Of the remaining 281 citizen scientist DBH measurements,
the difference between the citizen scientist measurement and the expert measurement was 1.2% on average (± 0.09% standard
error). These results show that citizen scientists collect valuable and accurate data, and demonstrating how important citizen
scientists are in large-scale scientific projects.
Overall, our results show that the survival rate of street trees in the City of Cambridge is on the higher end of scientific
estimates in other cities. Survival and growth rates for young trees are higher in residential areas compared to industrial and
office areas. Survival is also higher in areas of the city with higher owner occupancy rates, and growth rates are higher in areas
with higher population density. Combined, these results suggest that tree growth and survival improves when the trees are
cared for by the adjacent property owners or renters (i.e., abutters).
GOALS, OBJECTIVES, AND RESULTS
Project Background
With the increasing urbanization of the world’s population, the quality and sustainability of life in metropolitan areas is now a
national and global priority (UNFPA 2007). Managing urban forests to maximize ecosystem and community benefits is a priority
for many policy makers and city residents. Not only are urban trees aesthetically pleasing, but they also sequester carbon,
increase property values, improve health, and provide recreational opportunities (Ulrich 1984, Bolund & Hunhammar 1999,
Nowak & Dwyer 2000, Wolf 2003, Lovasi et al. 2008, Donovan & Butry 2009, Nowak et al. 2013, Kardan et al. 2015). By
providing shade and through evapotranspiration, urban trees have the potential to mitigate the urban heat island effect and
the effects of climate change.
Although the phrase “right tree, right place” is common in arboriculture, the scientific knowledge base for making the best
choices about which trees to plant where and how to best manage them is only just emerging. Urban environments are often
highly stressful for trees, and thus species choice and tree care is a non-trivial matter (Koeser et al. 2014, Maco & McPherson
2003). Urban trees need access to key resources such as water and nutrients successfully grow and survive. But they are often
challenged by multiple threats, including limited water availability due to drought and/or lack of permeable surface,
compressed soil with low oxygen content and little space for root growth, low soil nutrient content with high salt
concentrations, polluted air, various pests and pathogens, as well as physical damage by humans and storms (Craul 1994, 1985,
Foster & Blaine 1978, Jim 1993, 1998, Nowak et al. 2001, Britt & Johnston 2008). To further complicate matters, the nature of
these threats is continuously changing as global warming alters climatic patterns (Frumhoff et al. 2007).
The overarching goal of our study is to improve our understanding of the primary factors influencing variation in tree survival
and growth in urban environments. This type of information is vital for creating a more resilient urban forest that is able to
withstand the threats of an urban environment, and will help to ensure the best possible urban forest management practices
are being used. In our study we include various types of factors that may influence tree growth and survival, ranging from
biological (e.g. species and biological characteristics), to environmental (e.g. impervious surface surrounding tree, solar
radiation, tree well maintenance), to socioeconomic/ community factors (e.g. neighborhood level human population and
housing characteristics).
Results: City of Cambridge
The City of Cambridge maintains a spatially-explicit inventory of the more than 19,000 publicly-owned trees in the City
(https://www.cambridgema.gov/theworks/ourservices/urbanforestry/treeinventory). The inventory contains information
about the location, species identity, and size (diameter at breast height, or DBH) of each tree, as well as characteristics about
the site where the tree is planted, such as if there are wires overhead.
Between 2012 and 2015, citizen scientists participating in the “Exploring Boston’s Urban Forest” expedition visited and remeasured more than 4,000 trees in Cambridge’s tree inventory. Each tree was visited 1-3 times during the four-year period.
Combining these measurements with the original City of Cambridge tree inventory data resulted in 2-4 measurement time
points from which to assess tree survival and growth rates.
In addition to the tree measurements, we use spatial datasets from the City of Cambridge and the State of Massachusetts to
quantify environmental characteristics surrounding each tree, including the percent impervious surface surrounding each
tree, and the annual growing season light availability for each tree. We also characterize the socioeconomic conditions of
each neighborhood in the city using American Community Survey (ACS) data from US Census Bureau. A complete list of all
the biological, environmental, and socioeconomic/community variables used in this study are shown in Table 1.
Table 1. Biological, environmental and socio-ecological variables used in this study.
Component
Type
Biological
Data Source
Variable
Units
City inventory data
and citizen scientist
data
Environmental
City inventory data
and citizen scientist
data
Tree species
Initial tree size
Final tree size
Tree condition
Measurement date
Planting season
Root flare visible
Wires overhead
Sidewalk damage
Tree well maintenance
Impervious surface within 20m
radius of tree
Insolation
name
cm
cm
rating
date
date range
yes or no
yes or no
yes or no
yes or no
proportion
Variable
Type
Categorical
Continuous
Continuous
Categorical
Discrete
Discrete
Discrete
Discrete
Discrete
Discrete
Continuous
MJ m-2
Continuous
Population density
# people ha-1
Discrete
Housing density
# housing units ha-1
Discrete
percent
percent
dollars
year
percent
Continuous
Continuous
Continuous
Continuous
Continuous
percent
yes or no
yes or no
yes or no
yes or no
yes or no
Continuous
Discrete
Discrete
Discrete
Discrete
Discrete
Socioeconomic
Cambridge GIS
Database$
Massachusetts 2009
LiDar data#
US Census Data:
2009-2013 ACS
(census block group
level)^
Owner occupancy rate
Vacancy rate
Median household income
Median year housing built
Population with a college
degree (Associates or higher)
Population white
Cambridge GIS
Zoning, residential areas
Database$$
Zoning, commercial areas
Zoning, industrial areas
Zoning, open space areas
Zoning, other areas
$
Data from: www.cambridgema.gov/GIS/gisdatadictionary/Environmental.
$$
Data from: www.cambridgema.gov/GIS/gisdatadictionary/CDD.
#
Data from: www.mass.gov.
^
Data from: www.census.gov.
Tree Survival
Annual survival estimates were calculated separately for young trees (trees planted in 2007 or later that have a known plant
date) and old trees (trees established before 2007). Annual survival estimates were calculated using the following equation,
where t is the amount of time, in years, between the first recorded measurement and the last recorded measurement.
Annual Survival = (
!"#$%& !" !"##$ !" !"#$ !
!"#$%& !" !"##$ !" !"#$ !"#$
)!/! ,
The average annual survival rate of trees in the City of Cambridge is higher than in many other studies (Roman & Scatena
2011). We calculated the average annual survival rate for young trees to be 96.7%, meaning that only about 3% of young trees
die each year. The timing of planting (i.e., spring, summer, or fall) does not significantly impact annual survival rate of young
trees (ANOVA; df = 2, 18, F = 0.01, p = 0.995).
The average annual survival rate for old trees is 90.8%, meaning that about 9% of older trees die each year. Although the
average annual survival rates differ between old and young trees, the difference is not significant because the 95% confidence
intervals overlap (Table 2).
Table 2. Annual survival estimates for young trees and old trees.
Dataset
Range
initial DBH
(inches)
Estimated
annual survival
Young Trees
1.0–4.1
96.7% +/- 1.2%
93.8% – 99.5%
Old Trees
4.2–45.0
90.8% +/- 5.2%
79.0% – 102.7%
(mean +/- SE)
95% Confidence
Interval for annual
survival estimate
Norway maples comprise about one-third of the old trees in our survival study (499 trees out of 1571), and this species has the
lowest survival rate of all the species in our study (see the section below on species-specific trends). To test whether the low
survival rates of Norway Maple were driving the lower annual survival of old trees, we redid the analysis after removing Norway
Maples. The results did not change significantly. Without Norway Maples, the estimated annual survival rate of old trees was
91.7% +/- 6.4% (95% confidence interval = 77.2% – 106.2%).
To assess the impact of the various biological, environmental, and socioeconomic variables on survival, we used Cox
Proportional Hazards models. This type of model assesses the risk of mortality for different levels of each variable. The output
is a Hazard Ratio, which is scaled relative to one. Values above one mean that the variable increases the risk of death, and
values below one mean that the variable lowers the risk of death.
Five out of fifteen variables significantly influenced the risk of mortality for young trees (Figure 1a):
1.
2.
3.
4.
Young trees exposed to higher light conditions (Solar Insolation) were more likely to die.
Young trees in areas with higher median income levels were also more likely to die.
Young trees were less likely to die in areas with higher owner occupancy rates.
Young trees in residential or commercial zones were also less likely to die compared to young trees in other zones,
including offices, educational areas, government buildings, health care facilities, and transportation areas.
Nine out of sixteen variables significantly influenced the risk of mortality for old trees (Figure 1b):
1. Older trees with a higher proportion of impervious surface within 20m of the trunk were more likely to die.
2. Older trees were more likely to die in areas with higher housing densities or areas where a higher proportion of the
residents have a college degree.
3. Older trees in areas with higher population densities, higher vacancy rates, higher proportions of white residents, or areas
with housing that was built more recently were less likely to die.
5. As for the young trees, older trees in residential or commercial zones were also less likely to die compared to young trees
in other zones, including offices, educational areas, government buildings, health care facilities, and transportation areas.
Figure 1. Cox Proportional Hazards model results for a) young trees and b) old trees in the City of Cambridge, MA. Figures
demonstrate which biological, environmental, and socioeconomic factors influence the risk of a tree dying. In each figure,
squares that are to the right of the vertical line show that higher values of that variable increase the risk of a tree dying,
whereas points to the left of the vertical line mean that higher values of that variable reduce the risk of a tree dying. The
error bars are 95% confidence intervals. Points whose error bars overlap the vertical line (hollow points) are not significant.
Points whose error bars do not overlap the vertical line (filled points) significantly influence the risk of a tree dying. For
example, although the risk of a young tree dying tends to be higher for trees with a larger initial tree size (Initial DBH), the
pattern is not significant. However, young trees that experience higher light conditions (Solar Insolation) have a significantly
higher risk of dying than trees in lower light conditions.
Tree Growth
Tree growth rate was calculated separately for each surviving tree. Relative tree growth rate is calculated based on the tree
diameter, or Diameter at Breast Height (DBH) measured at two different time points. The equation to calculate relative growth
rate is as follows.
Relative Growth Rate =
DBH at time 2 − DBH at time 1
time 2 − time 1
We limit the dataset for growth to trees for which we have two measurement times, and each measurement is precise to one
tenth of an inch. The entire dataset for growth estimates consists of 1,845 trees. For all 1,845 trees, the average relative
growth rate is 0.26 inches per year (with a standard error of 0.007 inches per year).
Young trees have faster relative growth rates than old trees (two-sample, two-tailed t-test; df = 1355, t = -3.71, p = 0.0002;
Table 3). After being transplanted a tree may expend more energy towards developing their root system than growing their
stem and branches. We tested whether young trees grew significantly less in the first year after planting compared to later in
their development. Young trees tended to have faster growth rates after their first year of growth, but the difference was not
significant (two-sample, one-tailed t-test; df = 55.4, t = -1.58, p = 0.06). Young trees planted in the spring season have faster
growth rates than trees planted in the summer or fall (ANOVA; df = 2, 1357, F = 11.20, p < 0.0001; Table 4).
Table 3. Growth estimated for young and old trees. Values
followed by different letter are significantly different.
Dataset
Old Trees
Young Trees
Less than one year
More than one year
Number
of Trees
Relative Growth Rate
(inches / year)
481
0.22 ± 0.01 a
1,364
0.27 ± 0.01 b
Table 4. Growth rate estimates by planting
season, for young trees. Values followed by
different letter are significantly different.
Planting
Season
Number
of Trees
Relative Growth Rate
(inches / year)
Spring
667
0.34 ± 0.02 a
52
0.20 ± 0.05
Summer
443
0.26 ± 0.02 b
1,312
0.28 ± 0.01
Fall
250
0.24 ± 0.01 b
We used multiple linear regression to test the influence of the various biological, environmental, and socioeconomic variables
on tree growth. We found tree growth rates varied by species for both young and old trees (Table 5). Also, young trees that
were larger when they were planted (i.e., initial DBH), that were planted in areas with higher population densities, or that
were planted in residential zones had faster growth rates (Table 5). Tree size negatively influence growth rate for old trees,
and trees living in areas with higher vacancy rates also had lower growth rates. Among older trees, growth rates were higher for
trees living in areas with newer housing, or residential zones.
Table 5. Growth responses to biological, environmental, and demographic variables. Values are relative importance metrics
for terms included in the best-fit models; factors that do not contain a number were not included in the best-fit model.
Significant covariates have been colored green if the coefficient estimate is positive and red if the coefficient estimate is
negative. The sample sizes were 1350 for young trees, and 481 for old trees.
Variable
Young trees
Old trees
Species
0.053***
0.089***
Initial Tree DBH (inches)
0.041***
0.007*
Solar Insolation (MJ / m2)
0.002#
Impervious Surface (%)
Population Density (# people per ha)
0.004
0.008***
Housing Density (# houses per ha)
0.005#
Owner Occupancy Rate (proportion)
Vacancy Rate (proportion)
0.016*
Median Income ($)
Population Unemployed (proportion)
Population with College Degree (proportion)
0.006
Population White (proportion)
0.002#
Median Year Housing Built (year)
0.001#
0.014***
Zoning, Residential
0.010***
0.029***
Zoning, Commercial
Zoning, Industrial
0.001
Zoning, Open Space
Model Adjusted R2
***p<0.001, **p<0.01, *p<0.05, #p<0.1
0.106
0.150
Species-specific trends
Our survival dataset includes 65 different species, and our growth dataset contains 56 species. We were able to run analyses for
species with 50 or more individuals in the survival dataset, or 40 or more individuals in the growth dataset.
For young trees, species-specific annual survival rate estimates range from 92.3% per year for Apple, to 100% per year for
Callery Pear (Table 6). For the old trees, species-specific annual survival rate estimates range from 73.0% per year for Norway
Maple to 99.9% per year for Honeylocust (Table 6).
Among young trees, the fastest growing species are Elm sp. (hybrid) and Pin Oak (Table 7), although the growth rates of young
Swamp White Oak, Japanese Zelkova, American Elm, Callery Pear, and London Planetree are statistically equivalent. Among
young trees, apple has the lowest growth rate. The fastest-growing species among the old trees is Honeylocust, which grows
significantly faster than Red Maple or Norway Maple (Table 7).
Table 6. Species-specific annual survival estimates for young trees and old trees.
N
RANGE INITIAL
DBH (CM)
ESTIMATED ANNUAL
SURVIVAL (MEAN +/- SE)
1927
2.5–10.4
96.7 +/- 1.2
Pear, Callery
50
2.0–4.0
100% +/- 0%
Maple, Hedge
59
1.0–2.0
99.5% +/- 0.3%
Elm, American
78
1.0–4.0
99.1% +/- 0.6%
Oak, Pin
100
1.5–4.0
98.7% +/- 0.7%
Linden, Littleleaf
87
1.0–4.1
98.6% +/- 0.7%
Honeylocust
225
1.0–4.0
98.5% +/- 0.7%
Elm sp.
110
1.6–3.0
97.6% +/- 0.9%
Oak, Swamp White
68
1.3–4.0
97.6% +/- 1.4%
Maple, Red
181
1.0–4.0
96.1% +/- 1.2%
Lilac, Japanese Tree
68
1.7–3.0
96.0% +/- 2.2%
Zelkova, Japanese
71
1.8–4.0
95.1% +/- 2.1%
Cherry, Sargent
66
1.4–3.0
94.8% +/- 3.5%
Serviceberry
61
1.0–2.0
94.4% +/- 3.1%
Cherry, Japanese Flowering
73
2.0–4.0
94.2% +/- 2.9%
Ginkgo
62
1.7–4.0
92.9% +/- 3.8%
Planetree, London
84
1.0–4.0
92.8% +/- 3.1%
Apple
53
1.0–2.0
92.3% +/- 5.8%
All other species combined
417
1.0–4.0
94.1% +/- 2.3%
1571
4.2–45.0
90.8% +/- 5.2%
Honeylocust
81
5.0–23.0
99.9% +/- 0.1%
SPECIES
Young Overall
Old Overall
Oak, Pin
131
5.0–45.0
99.9% +/- 0.1%
Planetree, London
46
5.0–43.0
99.7% +/- 0.3%
Maple, Red
145
5.0–34.0
96.3% +/- 1.8%
Pear, Callery
178
5.0–21.0
96.0% +/- 2.7%
Linden, Littleleaf
174
5.0–39.0
90.3% +/- 6.9%
Maple, Norway
497
5.0–34.0
73.0% +/- 24.4%
All other species combined
294
4.2–37.0
93.3% +/- 3.9%
Table 7. Species-specific growth estimates for young and old trees.
Relative growth rate and relative GDD growth rate values are mean ±
standard error (SE). Within a column and age class, values with different
letters signify that the growth rates are significantly different. Within an
age class (young or old), species are sorted from fastest growth rate to
slowest growth rate.
SPECIES
Young Overall
N
RELATIVE GROWTH RATE
(INCHES PER YEAR)
1364
0.27 ± 0.01
Elm sp.
79
0.45 ± 0.04 a
Oak, Pin
80
0.44 ± 0.03 a
Oak, Swamp White
51
0.41 ± 0.04 ab
Zelkova, Japanese
56
0.36 ± 0.03 abc
Elm, American
57
0.36 ± 0.05 abc
Pear, Callery
45
0.35 ± 0.04 abcd
Planetree, London
55
0.28 ± 0.03 abcd
Honeylocust
161
0.27 ± 0.02 bcd
Maple, Red
120
0.23 ± 0.02 bcd
Cherry, Japanese Flowering
57
0.23 ± 0.04 bcde
Linden, Littleleaf
63
0.21 ± 0.04 bcde
Maple, Hedge
50
0.16 ± 0.03 cde
Lilac, Japanese tree
40
0.12 ± 0.04 de
Serviceberry
40
0.11 ± 0.04 de
Apple
44
0.01 ± 0.06 e
All other species combined
366
0.26 ± 0.02 bcd
481
0.22 ± 0.01
Old Overall
Honeylocust
54
0.26 ± 0.02 ab
Linden, Littleleaf
58
0.19 ± 0.02 bc
Pear, Callery
40
0.17 ± 0.03 bc
Maple, Norway
71
0.14 ± 0.02 c
Maple, Red
56
0.14 ± 0.02 c
All other species combined
202
0.30 ± 0.01 a
MEASUREMENTS IN OTHER CITIES IN THE GREATER BOSTON AREA
In 2015, the Boston urban forest program expanded into the cities of Boston, Chelsea, and Somerville. The goal in these cities
was to target a specific subset of 16 trees, in order to gather enough data on these species through time to do predictive growth
and survival analyses. Citizen-scientist volunteers measured a total of 383 trees in Boston, 256 trees in Chelsea, and 1,031 trees
in Somerville (Table 8).
Table 8. Number of trees measured in news cities in 2015 by Earthwatch citizen scientists.
Species
American Elm
American Linden
American Sycamore
Callery Pear
Cherry
Elm sp.
Honeylocust
Japanese Tree Lilac
Littleleaf Linden
London Planetree
Norway Maple
Pin Oak
Red Maple
Serviceberry
Silver Maple
Japanese Zelkova
Other
TOTAL
Boston
15
17
9
1
19
8
72
59
27
67
17
20
1
37
14
383
Chelsea
23
1
4
14
20
5
38
13
16
9
21
12
19
15
1
37
8
256
Somerville
24
29
60
70
109
40
60
74
76
84
56
74
62
37
43
51
21
1,031
No prior tree inventory records were available for the City of Boston. As such, the 2015 data collection was designed as a first
inventory, to be followed up at a future date. The range of sizes of trees measured for each species is shown in Figure 2. The
data collected by volunteers has been shared with the City Arborist of the City of Boston.
In Chelsea, Earthwatch citizen scientists visited the location of 220 tree wells that were measured in a 2009 tree inventory. Of
these trees, 164 were alive and 56 were dead, resulting in an annual survival rate of 95.2%. Earthwatch volunteers also
measured 36 trees that were planted between 2009 and 2014, and found that 8 of them had died. The sample size of planted
trees is not large enough to estimate an annual survival rate. The Earthwatch data from Chelsea has been shared with the Tree
Warden of the City of Chelsea.
Sufficient data was collected in Somerville to warrant detailed analyses across the city and by species. These analyses are
currently underway, and the data will be shared with the Planning & Zoning Department and the Parks & Open Space
Department by the end of April 2016.
Figure 2. Boxplot showing the range of sizes of the trees sampled by Earthwatch volunteers in Boston, 2016. The upper and
lower edges of the box represent the first and third quartile of the data, and the thick central line is the median value.
QUALITY CONTROL ANALYSES
Quality control analyses comparing the DBH measurements made by citizen scientists to measurements made by an expert
confirm that the citizen scientist data is largely accurate. After each of the initial citizen-scientist training sessions in 2014 and
2015, each group of volunteers collected data on the same 1-3 trees which an expert had also measured. In total across these
two years, citizen scientists made 296 measurements on 44 different trees. Of 296 DBH measurements, only 15 (5%) were large
errors, differing from the expert measurement by 10% or more. Of the remaining 281 citizen scientist DBH measurements, the
difference between the citizen scientist measurement and the expert measurement was 1.2% on average (± 0.09% standard
error).
SECTION TWO: Project Impacts
INCREASING SCIENTIFIC KNOWLEDGE
Total citizen science research hours
Each field day is approximately 6 hours long. Approximately 2 hours are spent training volunteers on the science of the project,
the research methods, and the specifics of data collection. Approximately 3 to 4 hours are spent collecting data. For each year,
we estimate the number of citizen science research hours as 6 hours multiplied by the number of participants.
Summary of Annual Fieldwork
2012
2013
2014
2015
2
6
11
9
# of trees surveyed
684
1,795
3,623
2,571
# of participants
43
85
220
195
# citizen science research hours
258
510
1320
1170
# of expedition teams
Peer-reviewed publications
1) Boukili, V.K.S, D.P. Bebber, T. Mortimer, G. Venicx, D. Lefcourt, M. Chandler, and C. Eisenberg. Testing the assumptions of
urban forest ecosystem services models with direct measurements of tree growth. Urban Forestry and Urban Greening, in
revision. *Earthwatch is acknowledged.
2) Boukili, V.K.S, C. Scott, G. Venicx, M. Chandler, D. Lefcourt, and C. Eisenberg. The Influence of Biological, Environmental,
and Socioeconomic Factors on Urban Tree Growth and Survival. In preparation. *Earthwatch is acknowledged.
Non-peer reviewed publications
1) Boukili, V., and G. Venicx. 2015. Earthwatch Urban Tree Program. Newsletter article for Urban Tree Growth and Longevity
website. http://www.urbantreegrowth.org/utgl-newsletter-articles.html.
2) Cambridge Urban Forest Management Plan, Section 2: Current State of the Urban Forest. In preparation.
3) Cambridge Urban Forest Management Plan, Section 4: Scientific Analysis of Current Trends in Growth and Survival of
Cambridge’s Street Trees and Management Recommendations. In preparation.
Presentations
1) Boukili, V., G. Venicx, D. Lefcourt, M. Chandler. 2014. A citizen science approach to monitoring street tree population
ecology in Cambridge, MA. Northeast Natural History Conference, Springfield, MA.
2) Boukili, V., G. Venicx, D. Lefcourt, M. Chandler. 2014. A citizen science approach to monitoring street tree population
ecology in Cambridge, MA. Ecological Society of America Annual Meeting, Sacramento, CA.
3) Boukili, V., G. Venicx, D. Lefcourt, M. Chandler. 2014. Variation in the survival and growth of urban trees in Cambridge, MA.
Earthwatch Summit, Cambridge, MA. (poster presentation)
4) Boukili, V.2014. Urban Forests. Instructional class for the public, Chelsea, MA.
5) Boukili, V. 2016. Exploring Boston’s Urban Forest: Project Wrap Up. Earthwatch Mission Meeting, Boston, MA.
MENTORING
Community outreach
Name of school,
organization, or group
City on a Hill Charter School
(Roxbury, Boston, MA)
Education
level
High School
students
Participants local
or non-local
Local
Cambridge Science Club for
Girls (Cambridge, MA)
High School
students
Local
Groundwork Somerville
(Somerville, MA)
High School
students
Local
Cambridge Green Sense
(Cambridge, MA)
High School
students
Local
Stonehill College summer
program (Easton, MA)
High School
students
Local
Details on contributions/ activities
Led field day activity with math students (2
days). Training on the importance of urban
forests, identifying tree species, and making
scientific measurements.
Led field day activity with students in summer
program (1 day). Training on the importance of
urban forests, identifying tree species, and
making scientific measurements.
Led field day activity with students in summer
program (4 days). Training on the importance of
urban forests, identifying tree species, and
making scientific measurements.
Led field day activity with students in summer
program (1 day). Training on the importance of
urban forests, identifying tree species, and
making scientific measurements.
Led field day activity with students in summer
program (1 day). Training on the importance of
urban forests, identifying tree species, and
making scientific measurements.
PARTNERSHIPS
Support Type(s)1
Partner
Years of Association
(e.g. 2006-present)
2012–2016
City of Cambridge (David Lefcourt, City
Funding, data,
Arborist, and Owen O’Riordan, DPW
logistics, collaboration
Commissioner)
City of Chelsea (Andrew DeSantis, Tree Warden
Funding, data, logistics
2014–2016
& Assistant Director of Public Works)
City of Somerville (Rachel Kelly, Green
Data, logistics
2014–2016
Infrastructure Planner)
City of Boston (Gregory Mosman, City Arborist)
Data, logistics
2014–2016
1
Support type options: funding, data, logistics, permits, technical support, collaboration, academic support, cultural support,
other (define)
CONTRIBUTIONS TO MANAGEMENT PLANS OR POLICIES
2
Plan/Policy Name
Type2
Urban Forest
Management Plan,
City of
Cambridge, MA.
Management
plan
Level of
Impact3
Local,
regional
New or
Existing?
New
Primary goal of
plan/policy4
Natural resource
conservation/ other
(understand how
current practices are
working, and making
improvements)
Stage of
plan/policy5
In progress
Description of
Contribution
Boukili wrote the draft
of the plan, and
collaborators from
Earthwatch and the City
of Cambridge provided
edits, revisions, and
recommendations
Type options: agenda, convention, development plan, management plan, policy, or other (define)
Level of impact options: local, regional, national, international
4
Primary goal options: cultural conservation, land conservation, species conservation, natural resource conservation, other
5
Stage of plan/policy options: proposed, in progress, adopted, other (define)
3
ECOSYSTEM SERVICES
☐Food and water
☒Flood and disease control
☒Spiritual, recreational, and cultural benefits
☒Nutrient cycling
Urban trees provide various ecosystem services, including sequestering carbon, diminishing air pollution, mitigating flooding
potential, increasing property values, improving health, and providing recreational opportunities. They are also aesthetically
pleasing. Larger trees provide significantly more ecosystem service benefits than small trees. Our research findings will provide
a better understanding of urban tree growth and survival. By informing land managers, we can improve the likelihood that these
trees will survive until maturity and grow to their full potential, and accordingly provide increasingly more ecosystem service
benefits over time.
RESEARCH PLAN UPDATES
• Have you added a new research site of has your research site location changed? No.
• Has the protected area status of your research site changed? No.
• Has the conservation status of a species you study changed? No.
• Have there been any changes in projects scientists or field crew? No.
Provide details on any changes to your objectives, volunteer tasks, or methods, include reason for the change.
I submitted the last Earthwatch proposal for this project in 2015, which included an ambitious three-year plan with three
primary objectives. Due to lack of funding and the retirement of this project, I was not able to make any progress towards
objective 1 (dendrochronology study), or objective 3 (physiological study). The progress towards objective 2 (regional study)
was also limited.
ACKNOWLEDGEMENTS
Funding to carry out this project was provided by the generosity of the Borun Family Foundation, the Goldring Family
Foundation and Ernst and Young, LLC. We are also grateful to the City of Cambridge for providing funding to create an Urban
Forest Management Plan, and to the City of Chelsea for funding the creation of an instructional urban forestry class. This
project would not have been possible without the support of Earthwatch staff, particularly Gitte Venicx, Mark Chandler, and
Cristina Eisenberg, as well as Earthwatch citizen scientists, including employees of Ernst and Young, who collected the majority
of the data for our tree censuses. We are also grateful to Jessi Flynn for making additional tree measurements, and to Corwin
Scott for helping with database management and i-Tree analyses. Scientific discussions with Lara Roman, Lucy Hutyra, Pamela
Templer, and Andrew Reinmann were particularly helpful in focusing and refining our research questions.
LITERATURE CITED
• Bolund, P., and S. Hunhammar. 1999. Ecosystem services in urban areas. Ecol. Econ. 29: 293–301. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S0921800999000130.
• Britt, C., and M. Johnston. 2008. Trees in Towns II: A new survey of urban trees in England and their condition and
management. London, UK.
• Craul, P. J. 1985. A description of urban soils and their characteristics. J. Arboric. 11: 330–339.
• Craul, P. J. 1994. Urban soils: an overview and their future. In G. W. Watson and D. Neely (Eds.) The Landscape Below
Ground. International Society of Arboriculture, Savoy, IL.
• Donovan, G. H., and D. T. Butry. 2009. The value of shade: Estimating the effect of urban trees on summertime electricity
use. Energy Build. 41: 662–667.
• Foster, R. S., and J. Blaine. 1978. Urban tree survival: trees in the sidewalk. J. Arboric. 4: 14–17.
• Frumhoff, P. C., J. J. Mccarthy, J. M. Melillo, S. C. Moser, and D. J. Wuebbles. 2007. Confronting climate change in the U.S.
northeast. Synthesis report of the northeast climate impacts assessment (NECIA). Cambridge, MA.
• Jim, C. Y. 1993. Soil compaction as a constraint to tree growth in tropical and subtropical urban habitats. Environ. Conserv.
20: 35–49.
• Jim, C. Y. 1998. Impacts of intensive urbanization on trees in Hong Kong. Environ. Conserv. 25: 146–159.
• Kardan, O., P. Gozdyra, B. Misic, F. Moola, L. J. Palmer, T. Paus, and M. G. Berman. 2015. Neighborhood greenspace and
health in a large urban center. Sci. Rep. 5: 11610. Available at: http://www.nature.com/doifinder/10.1038/srep11610.
• Koeser, A. K., E. F. Gilman, M. Paz, and C. Harchick. 2014. Factors influencing urban tree planting program growth and
survival in Florida, United States. Urban For. Urban Green. 13: 655–661. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S1618866714000739.
• Lovasi, G. S., J. W. Quinn, K. M. Neckerman, M. S. Perzanowski, and A. Rundel. 2008. Children living in areas with more
street trees have lower prevalence of asthma. J. Epidemiol. Community Helath 62: 647–649.
• Maco, S. E., and E. G. McPherson. 2003. A practical approach to assessing structure, function, and value of street tree
populations in small communities. J. Arboric. 29: 84–97.
• Nowak, D. J., and J. F. Dwyer. 2000. Understanding the benefits and costs of urban forest ecosystems. In Kuser (Ed.)
Handbook of urban and community forestry in the North East. pp. 11–25, Kluwer Academic/ Plenum Publishers, New York,
NY.
• Nowak, D. J., E. J. Greenfield, R. E. Hoehn, and E. Lapoint. 2013. Carbon storage and sequestration by trees in urban and
community areas of the United States. Environ. Pollut. 178: 229–36. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/23583943 [Accessed January 24, 2014].
• Nowak, D. J., M. H. Noble, S. M. Sisinni, and J. F. Dwyer. 2001. People & trees: assessing the US urban forest resource. J.
For. March: 37–42.
• Roman, L. A., and F. N. Scatena. 2011. Street tree survival rates: meta-analysis of previous studies and application to a
field survey in Philadelphia, PA, USA. Urban For. Urban Green. 10: 269–274. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S1618866711000422 [Accessed March 21, 2014].
• Ulrich, R. S. 1984. View through a window may influence recovery from surgery. Science (80-. ). 224: 420–421.
• UNFPA. 2007. State of the world population 2007: unleashing the potential of urban growth. New York, USA.
• Wolf, K. L. 2003. Public respone to the urban forest in inner-city business districts. J. Arboric. 29: 117–126.
Earthwatch Institute
114 Western Avenue
Boston, MA 02134
U.S.A.
1.800.776.0188
[email protected]
earthwatch.org
facebook.com/earthwatch
twitter.com/earthwatch_org