A Market Analysis of Opt-In Tree Planting and Rain Barrel

How Do Tree and Rain Barrel Programs Work?

The data analyzed here represent a selecon of programs where

residents acvely opted in and parcipated. Four types of tree

planng programs were analyzed: I) Street trees: Balmore City Rec-

reaon and Parks (BCRP) plants street trees. This report analyzes only

those street trees planted by the BCRP sta in response to resident-

placed requests, which does not include all street trees planted by

BCRP or trees planted in other places such as parks.

Why Are Trees and Rain Barrels Important?

The Balmore Sustainability Plan

established a goal to increase tree

canopy cover from 27%

to 40% by the year 2037

. TreeBalmore is

helping to grow the urban forest in partnership with other groups.

Rain barrels reduce the volume of water owing o of buildings and

into storm drains. They reduce polluon and prevent erosion

. Blue

Water Balmore (BWB) has a number of environmental educaon

and storm water management programs. These include three types

of rain barrel installaon programs.

Figure 1. Resident-requested planted street trees and giveaway trees for residenal properes (le), rain barrels installed via Blue Water Bal-

more’s dierent programs (right). Data are from year 2008 to 2012. Trees and rain barrels beyond Balmore City boundaries are not shown.

A Market Analysis of Opt-In Tree Planng and Rain Barrel

Installaon in Balmore, MD, 2008—2012

This report is movated by the overarching queson “Where were

trees planted and rain barrels installed, as part of opt-in programs

in relaon to Balmore’s priories?” In parcular:

1. Who parcipated in planng and installaon programs in terms

of available space?

2. How did parcipaon vary by TreeBalmore priories?

3. Where are there opportunies for tree planng in terms of availa-

ble space and priories?

Note too that other organizaons plant street trees in Balmore City.

II) Giveaway trees: TreeBalmore and BWB give trees to residents

at farmers’ markets, school planngs, and other events to be planted

on private property (Figure 1, le). This includes trees distributed by

TreeBalmore to partner organizaons. III) Volunteer planngs:

Trees planted by larger and highly professionalized organizaons

using volunteers. These groups include the Alliance for the Chesa-

peake Bay, Balmore Orchard Project, Balmore Tree Trust, BWB,

Jones Falls Watershed Associaon, and the Parks & People Founda-

on. IV) Neighborhood planngs: Trees planted by local, smaller

community organizaons, business improvement districts and faith-

based groups including Downtown Partnership, Druid Hill Community

Development Group, Ellerslie Public Housing, Episcopal Community

Church, Gary Leeron, Labyrinth Sacred Space, Midtown Community

Benets District, Morgan State Delta Sigma Theta Sorority, Mt

Vernon Business Associaon, New Cathedral Cemetery, Reservoir Hill

Improvement Council, The Samaritan Women’s Rain Garden, and

Windsor Hills Elementary.

This report examines 3 types of rain barrel installaon programs

(Figure 1, right): BWB’s auditor visits a residence and determines site

suitability with the resident and makes a recommendaon whether

to install a rain barrel. BWB acts as wholesaler of the rain barrel and

associated equipment. Note that water audit and workshop parci-

pants can purchase the materials at a reduced cost. BWB hosts edu-

caonal workshops either at their headquarters (walk ups wel-

come) or in dierent neighborhoods at the request of residents or

community groups.

What are Geodemographic Analyses?

To answer the queson “Who parcipated in planng and installaon

programs?” we used geodemographic market analyses. Geodemo-

graphic market analysis is the process of classifying neighborhoods

into groups based on a combinaon of demographic, socioeconomic

and lifestyle characteriscs. These market groups do not correspond

neatly to a single characterisc such as race/ethnicity, or the age

distribuon within a neighborhood. Instead, each market group de-

scribes a range of useful traits among people clustered in similar geo-

graphical locaons, such as consumer behaviors and preferences and

lifestyles. This analysis uses Environmental Systems Research Ins-

tute’s (ESRI) Tapestry Segmentaon system, which includes 65 mar-

ket segments that are consolidated into 12 Tapestry Groups to pro-

vide proles at the US census block level

. Figure 2 shows how these

categories are distributed across the city geographically (le), and

quantavely, (right).

The tree and rain barrel programs strive to work throughout Bal-

more City (Figure 1, previous page). Balmore City is comprised of

11 Tapestry Groups. The three most common Tapestry Groups are

Metropolis (comprising 42% of all households), Solo Acts (15%), and

Tradional Living (13%; Figure 2). Together these three Tapestry

Groups contain 70% of all households living in Balmore City. Below,

some of the dominant and dening characteriscs of each of these

three Tapestry Groups are highlighted from ESRI’s Tapestry classica-

on system.

More informaon on all Tapestry Groups can be found in the Refer-

ence Guide: hp://www.esri.com/library/brochures/pdfs/tapestry-

segmentaon.pdf

Figure 2. ESRI’s geodemographic market segments called Tapestry Groups have a geographic (le) and quantave (right) distribuon.

Who Are the Tapestry Groups in Balmore?

As the ESRI Tapestry Reference Guide explains:

Metropolis: They live in older, single-family homes or row houses

built in the 1940s or earlier. Workers in most of the Metropolis seg-

ments commute to service-related jobs. The Metropolis group re-

ects the segments’ diversity in housing, age, and income. For exam-

ple, ages among the segments range from Generaon Xers to rer-

ees; households include married couples with children and single

parents with children. Employment status also varies from well- edu-

cated professionals to unemployed. Their lifestyle is also uniquely

urban and media oriented.

Solo Acts: residents are singles who prefer city life. Many are young,

just starng out in more densely populated US neighborhoods; oth-

ers are well-established singles who have no home ownership or

child-rearing responsibilies. Residents of this group tend to be well-

educated, working professionals who are either aending college or

already hold a degree. Their incomes reect their employment expe-

rience, ranging from a low median of $44,601 among the newest

households to approximately $93,899 among established singles.

Home ownership is at 28 percent. Contrary to modern migraon

paerns that ow away from the largest cies, Solo Acts’ residents

are moving into major cies. With considerable discreonary income

and few commitments, their lifestyle is urban, including the best of

city life—dining out, aending plays and concerts, and vising muse-

ums—and, for a break from constant connecvity, extensive travel

domescally and abroad.

Tradional Living: The households in Tradional Living convey the

percepon of real middle America—hardworking, seled families.

The group’s higher median age of 38.2 years also conveys their

lifestage—a number of older residents who are compleng their

child-rearing responsibilies and ancipang rerement. Many sll

work hard to earn a modest living. They typically own single-family

homes in established, slow-growing neighborhoods. They buy stand-

ard, four-door American cars, belong to veterans’ clubs and fraternal

organizaons, take care of their homes and gardens, and rely on tra-

dional media such as newspapers for their news.

Key Terms Tree Canopy: PROW and Residenal Lands

Geodemographic market segmentaon: the process of classifying

neighborhoods into categories based on a combinaon of

demographic, socioeconomic and lifestyle characteriscs. Tapestry

Groups are an example of a market segment.

Tree Canopy (TC): Tree canopy is the layer of leaves, branches, and

stems of trees that cover the ground when viewed from above.

Exisng TC: The amount of urban tree canopy present when viewed

from above using aerial or satellite imagery.

Possible TC: Asphalt or concrete surfaces, excluding roads and build-

ings, and grass or shrub areas which are theorecally available for the

establishment of tree canopy.

Not Suitable: Areas where it is highly unlikely that new tree canopy

could be established (primarily buildings and roads).

Odds rao: A measure of associaon where 1 indicates that parci-

paon is proporonal to the number of households in that Tapestry

Group. A value of 2 would indicate that twice as many trees were

registered than if giveaways were equitability distributed by Tapestry

Group. See Figure 4 (over).

To answer the queson “How did parcipaon vary by available

space and priories?” we report results for both street trees in the

public right of way (PROW) and for private residenal lands. Within

the PROW, the amount of exisng and possible tree canopy and areas

not suitable for planng were calculated within each Tapestry Group

(Figure 3). For exisng PROW, High Society block groups have the

most tree canopy (54%), while Family Portrait has the least tree can-

opy (12%). For possible PROW, High Society block groups have the

least area (14%), while Factories & Farms have the most area (34%).

These analyses were repeated for residenal properes too. Residen-

al properes include apartments, condominiums, row homes, single

-family and two-family homes. Exisng tree canopy on residenal

properes was highest among High Society block groups (65%) and

lowest in Scholars & Patriots areas (18%). The Tapestry Groups with

the greatest opportunies for addional canopy were Factories &

Farms and Tradional Living. Each had 46% possible tree canopy.

High Society block groups had the least room for addional possible

tree canopy on residenal lands (26%). There were ample opportuni-

es for addional tree canopy on residenal properes and in the

PROW in every Tapestry Group (Figure 3A & C).

High Society

Metropolis

Upscale Avenues

Senior Styles

Solo Acts

Scholars & Patriots

High Hopes

Global Roots

Family Portrait

Tradional Living

Factories & Farms

Public Right of Way,

top bar

Residenal Parcels,

boom bar

Figure 3. Exisng and possible tree canopy in the PROW and on residenal parcels, per Tapestry Group (A), exisng tree canopy (B) an d

possible tree canopy (C) on residenal parcels only by Tapestry Group. Colors correspond to the Tapestry Groups. The height of each block

group indicates the exisng (B) or possible tree canopy (C). Canopy calculaons were made possible with the combinaon of Balmore City’s

parcel database, and the 3-foot freely available land cover data created by the Spaal Analysis Lab at the University of Vermont.

Exisng Tree Canopy on

Residenal parcels by

Tapestry Group

Possible Tree Canopy on

Residenal parcels by

Tapestry Group

B C

Measuring Parcipaon Rates with Spaal Analyses and Tapestry

Addresses represenng the four types of tree programs, and three rain barrel programs described on page 1 were converted to points using a

geographic informaon system (GIS), or computer mapping soware (Figure 1). These analyses assume that the tree was planted or the rain

barrel was installed at that locaon, but the actual locaon is unknown. These points were analyzed within each market segment as dened by

ESRI’s Tapestry Groups (Figures 2, 3B & C) by calculang odds raos. Figure 4 describes how odds raos were calculated with points repre-

senng the assumed planng locaon of trees and installaon sites of rain barrels within Tapestry Groups, using the cous example of

“Phake City”

. Actual odds raos for all trees and rain barrels are shown in Figure 5.

Figure 4. An odds rao (OR) is a measure of

associaon where a value of one indicates an equal

proporon eect in a single Tapestry Group. Odds

raos greater than one indicate higher odds of

occurrence, and a value less than one means there

is a lower odds. Phake City

has 3 types of market

segments, the number of households are shown in

each market. There are 100 households in total. The

hypothecal urban forestry program gave out 10

trees last year, reaching 10% of all households. If

each type of market segment were reached equally

(ie. proporonally to the number of households in

that Tapestry Group), then the white segment

would have 1 tree, the gray would have 5 trees, and

the remaining 4 would be in the darkest gray. But

that is not the case. Instead households in the white

market segment received 2 (twice as many, OR = 2),

the gray received 2 trees (40% of expected, OR= .4),

and 6 trees went to the households in the darkest

gray segment (50% above expected, OR = 1.5).

Odds Ratios (OR) & 95% Confidence Intervals for Tree & Rain Barrel Programs, by Tapestry Group

Figure 5: Parcipaon in tree (A) and rain barrel (B) programs varies by Tapestry Group. Tree giveaways were most popular in the High

Society areas (OR = 5.25), volunteer planngs were most popular in Farms & Factories neighborhoods (OR = 4.42), and neighborhood planngs

were most popular in Family Portrait areas (OR = 3.84). BWB’s water audits were most popular in Upscale Avenue & High Society neighbor-

hoods (OR = 5.34 & 5.22), wholesales and workshop rain barrels were most popular in Upscale Avenues neighborhoods (OR = 4.20 & 5.93) .

Figure 7 (le) shows how possible tree canopy re-

lates to market segments and planng priories.

This table can be used in several ways to create mar-

ket strategies based upon combining UTC Assess-

ments (available land), UTC Priorizaons (priority

areas) and UTC Market Analyses (parcipaon rates

among Tapestry Groups). The following are some

illustraons:

1) Current programs are most successful with High

Society and Upscale Avenues (Figure 5A), but there is

very lile addional space in High priority areas with-

in those Tapestry Groups (Figure 3A & C). Current

programs might connue in these two Tapestry

Groups, but only in those neighborhoods that are a

high priority.

2) Hit the ‘sweet spot’ within the doed red line

where parcipaon rates are currently very low but

extensive areas are available in High to Medium pri-

ority areas. New markeng strategies may be needed

to increase parcipaon for these Tapestry Groups to

achieve the City’s Urban Tree Canopy Goal with trees

in the highest priority areas.

3) Developing new markeng strategies for areas

with low parcipaon rates may not be useful be-

cause some of these High priority areas have lile

available land for tree planng, such as Tradional

Living and Family Portrait areas (Figure 3A & C).

Other combinaons and strategies are possible.

High Society

Upscale Avenues

Metropolis

Solo Acts

Senior Styles

Scholars & Patriots

High Hopes

Global Roots

Family Portrait

Tradional Living

Undened

0 to 2 acres

2 to 150

150 to 350

350 to 3,000

Possible Tree Canopy by

Tapestry Group and Planting Priority

Figure 6: Trees planted in the Fall of 2013 are mapped along with UTC priories (A)

and the organizaons that planted the trees (B). Some organizaons work across

priority areas like Parks & People Foundaon, while other groups work predomi-

nantly in High priority areas like the Balmore Tree Trust and TreeBalmore.

Urban Tree Canopy Priorizaon

An urban tree canopy priorizaon map was created for Balmore based upon a stakeholder engagement process lead by TreeBalmore and

its partners in collaboraon with the USDA Forest Service in 2012 (see Figure 6 and reference #5 for more details). Examples of priories in-

clude stormwater and air quality migaon, and environmental equity.

Tapestry Group

High Medium/ Medium Medium/ Low

High

UTC Priority

Summary UTC Prioritization Map

How Does Possible Tree Canopy Relate to Market Segments and Planng Priories?

References

Dexter H. Locke MESc. and J. Morgan Grove, PhD.

Quesons about this report can be addressed to:

dexter.locke@gmail.com and morgangrove@fs.fed.us

For more informaon see: hp://www.treebalmore.org/

Prepared by:

Findings

1. Balmore Sustainability Plan (2009).

2. O’Neil-Dunne, J.P.M (2009). A report on the City of Balmore’s

Exisng and Possible Tree Canopy.

3. Blue Water Balmore (2015).

4. ESRI (2010) Tapestry Segmentaon Reference Guide - Esri.

5. Adapted from Locke, D.H. (2014): Phake City. gshare.

6. Locke, D. H., Grove, J., Galvin, M. F., O’Neil-Dunne, J. P. M., &

Murphy, C. (2013). Applicaons of Urban Tree Canopy

Assessment and Priorizaon Tools: Supporng Collabora-

ve Decision Making to Achieve Urban Sustainability Goals.

Cies and the Environment (CATE), 6(1).

6. Note that most of this report is adapted from Locke, D.H., &

Grove, J.M. (2014). Doing the Hard Work Where it’s Easi-

est? Examining the Relaonships Between Urban Greening

Programs and Social and Ecological Characteriscs. Applied

Spaal Analysis and Policy. doi:10.1007/s12061-014-9131-1

Queson 1: Who parcipated in planng and installaon programs

in terms of available space?

 Tree giveaways were most popular in the High Society and Up-

scale Avenues neighborhoods. These are the two most auent

market segments and collecvely comprise about 3% of Bal-

more City’s households. Tree planng program parcipaon co-

varied with the exisng tree canopy in a potenally counterintui-

ve manner—the more canopied neighborhoods, which have less

available planng space (Figure 3A), also tended to have the

greatest parcipaon in tree planng programs (Figure 5A).

 Volunteer planngs were most popular in Farms & Factories

neighborhoods, the least auent market segment, which makes

up less than half a percent of households (Figure 5A).

 Neighborhood planngs were most popular in Family Portrait

neighborhoods, but were generally low overall (Figure 5B).

 BWB’s water audits were most popular in Upscale Avenue & High

Society neighborhoods.

 Wholesale and workshop rain barrels were dominated by house-

holds in the Upscale Avenues, High Society, and Tradional Living

market segments. Notable excepons include unpopular work-

shop programs in the High Society areas, and higher than ex-

pected parcipaon in the wholesale program by Scholars & Pa-

triots households (Figure 5B).

Queson 2: How did parcipaon vary by the City’s priories (Figure

6A & B)?

 TreeBalmore and the Balmore Tree Trust planted only in High

priority areas in the fall of 2013.

 Parks and People Foundaon planted across all priority areas.

 Blue Water Balmore planted in all priority areas except for High

priority areas.

Queson 3: Where are there opportunies for tree planng in terms

of available space and priories?

 There is ample opportunity for addional tree planng in every

market segment (Figure 3A & C). This is especially true for the

street trees in the Farms and Factories and Family Portrait market

segments where more than 30% of the PROW is possible tree

canopy. On private residenal lands, the greatest opportunies

by area can be found in the High Hopes, Senior Styles, Global

Roots, Factories and Farms, and Tradional Living market seg-

ments where more than 35% of land area is possible tree canopy.

 A plurality of Balmore City‘s households are classied as Me-

tropolis (42%) and they are a media-oriented segment. Since the

TreeBalmore outreach campaign has used more digital media

since 2012, follow up analyses could examine the increase of

programs specically in this market segment because parcipa-

on was previously low (Figure 5). There are extensive areas of

available land to plant in Metropolis neighborhoods (Figures 3A &

C, 7).

 While more resources may be needed overall to reach the tree

canopy goals

to achieve the environmental and social benets of

trees, there may be alternave, more cost-eecve ways of

reaching these dierent social groups in dierent priority areas

through more eecve markeng (Figure 7, Suggeson #2).

 The Green Paern Registry can be used for further tracking and

analyses. New capabilies are being added to the Green Paern

Registry to track tree plants by locaon, and by organizaon or

household. The Green Paern Registry is hosted by the Bal-

more Neighborhood Indicators Alliance (water.bniaj.org/

map/). Green Paern Registry supports the Growing Green Ini-

ave and City of Balmore’s Green Paern Book, a guide to help

convert vacant lots into mul-funconal, sustainable, and resili-

ent landscapes.

 Adopon rates were highest in areas that already had the high-

est rates of canopy cover and were the most auent. The ’sweet

spot’ areas containing the most available land for tree planng in

the highest priority areas had low rates of parcipaon. Tree

programs may consider which messages, delivered with which

messenger, will be most successful in these market segments?

For instance, it may be possible that outreach could be more

eecve when a locally appropriate type or combinaon of mes-

sages–content or the “what”–and messengers–means of mes-

sage delivery or the “how”–are matched to the needs and per-

cepons of people in dierent market segments.

 Finally, in addion to the programs examined here, there are

several other urban forestry programs and projects carried out

by TreeBalmore and their partners. Each of these eorts helps

reach the 40% tree canopy. For example, the Parks & People

Foundaon and Blue Water Balmore plant trees on schoolyards

because of their ample space, the availability of student volun-

teers, and relavely low costs.

Suggesons