Child pages
  • Kaskey.FinalAssignment

Lauren Kaskey

UEP 231: GIS Final Paper

December 15, 2013

Opportunities for Educational Urban Agriculture in Boston, MA

The City of Boston’s Office of Food Initiatives (OFI) has partnered with the Boston Redevelopment Authority (RDA) to update the city’s zoning laws for urban agriculture. The new Urban Agriculture zoning laws are meant to streamline the process of establishing a farm within city limits. These regulations have the potential to increase access to fresh and healthy food throughout the city, particularly in neighborhoods experiencing a disproportionate level of blight. In 2013, OFI sponsored a team of Tufts graduate students in surveying opportunities for urban agriculture in Boston. For this project, the same methodology and criteria was used to identify opportunities for Urban Agriculture in Boston, with a few modifications. Rooftop farming is also addressed in Article 89, and a survey of potential rooftops is included in this project.

A “Prime” Parcel for Urban Agriculture:

  1. Is over 10,000 square feet.
  2. Is vacant.
  3. Has low levels of shade.
  4. Is less than a 20% slope.

It has been shown that involvement in vegetable cultivation can increase children’s acceptance of fruits and vegetables. [i] , [ii] , [iii] This highlights public health potential for including an educational element to farms throughout the city, and especially in the neighborhoods experiencing health disparities such as high levels of chronic disease and childhood obesity. When creating plans for neighborhood development, organizations might want to prioritize vacant land plots and eligible rooftops that are within walking distance, or a PedShed, of a Boston Public School. As neighborhoods in Boston invest in developing blighted properties, information about their potential as an urban farm, and especially one with an educational component, will help them make decisions about the best use for their land.

There are few parcels that meet the “prime farming” criteria. However, most of these parcels are located in walking distance to a school. Additionally, there are quite a few rooftops that meet the size requirement for profitable agriculture. More research should be done to evaluate the potential for rooftop farming in Boston.  These parcels and rooftops, if developed into farms, have the potential to partner with school groups to support healthy initiatives in Boston Schools.



Data Sets:

Projected Coordinate System: NAD_1983_StatePlane_Massachusetts_Mainland_FIPS_2001


MassGIS ( ):

  • Boston Parks and Recreation: Protected and Recreational Open Space Updated 9/20/2013
  • MassDOT: EOTRoads_35 Updated 4/3/2012
  • MBTA Rapid Transit: MBTA_ARC Updated 7/2/2013
  • Trains (and MBTA Commuter Rail): TRAINS_ARC Updated 4/2008
  • MassDEP Hydrography (1:25,000): HYDRO25K_POLY, HYDRO25K_ARC Updated 3/2010
  • DEP Wetlands (1:12,000): wetlandsdep_arc Updated 1/2009
  • Zones: Zones_35 Updated 8/2007
  • Mass. Department of Agricultural Resources: Farmers Markets Updated 7/16/2003


Boston Data Portal ( ):

  • Schools 2012 – 2013 Updated 10/1/2012
    • Spreadsheet cleaned for use in ArcMap.
    • Geocoded using XY data.


  • New layer created, projected to Data Frame.
  • Community Gardens Updated 1/9/2013
    • Spreadsheet cleaned for use in ArcMap.
    • Geocoded using XY data.
    • New layer created, projected to Data Frame.
  • School Gardens Updated 4/11/2013
    • Spreadsheet cleaned for use in ArcMap.
    • Geocoded using XY data.
    • New layer created.
  • Urban Farms Updated 1/9/2013
    • Geocoded using XY data.
    • New layer created, projected to Data Frame.

Tufts University:

  • Parcels: Boston 2012 Parcels with Condo Fix
    • Projected to appropriate data frame.
    • Joined to Assessor_Data 2012 using preset join.


  • Trees: City of Boston DOIT Plenimetric 2010
    • Projected to appropriate data frame.
  • Buildings: City of Boston DOIT Plenimetric 2010
  • Sidewalks: City of Boston DOIT Plenimetric 2010
  • Elevation: Massachusetts_DEM_1_arc_second_NHD_projected
  • Grove Hall: BRA BostonNeighborhoods00

Statewide data was clipped using the BRA BostonWardsPrecincts00.shp retrieved from the Tufts Depository.

Determining Appropriate Vacant Land:

From Parcels: Vacant land greater than 10,000 square feet.

  • Select by attribute -> AV_BLDG=0 -> Create new layer, Empty_Parcels.

  • Select Empty_Parcels. -> select by location to “remove from currently selected features in” “Empty_Parcels” Source layer: “Boston Parks and Recreation,” spatial selections “Intersect source layer feature


  • Use the same process for:
    • Wetlands
    • Commuter Rail
    • Roads
    • Water
    • Open_Space
    • MBTA


  • We want to consider contiguous parcels to be one potential site, so we will dissolve parcels and calculate the total area of the site to determine which potential parcels are 10,000 sqft or more.


  • Combining sites using Dissolve tool:
    • Toolbox: Data Management Tools: Generalization: Dissolve: Parcels
    • New column named sq_ft
    • Calculate area of all parcels
  • Calculate geometry -> select by attribute: sqft >= 10000
  • Create layer of contiguous parcels > 10,000 sqft called Parcels_Contiguous
  • Determine which parcels intersect: Select by location: Add to selected: Parcels intersect with Parcels_Contiguous
  • Create new layer-> Parcels_and_Cont
  • Calculate Geometry-> Create layer from all >10k sqft


Although the previous team excluded land that was 20% or less non-impervious, this project kept these spaces in analysis. Many Boston farms will be using raised beds, which can be erected on parking lots and other spaces with concrete or other impervious cover. In fact, putting a raised bed on a space that is currently asphalt seems to be an effective way to prevent the heat magnet phenomenon found in so many cities with parking lots. Additionally, Urban Farmers are historically creative with the farm sites, and it will be left to their judgment whether or not an impervious surface is appropriate for farming.


Eliminating those with steep Slope:

  • Create a slope raster of the Digital Elevation Model (DEM) of Boston.

  • Use Spatial Analyst to determine elevation of each parcel:
    • Spatial Analyst-> Zonal -> Zonal Statistics as Table -> select parcel layer as input, zone field = Parcel ID, DEM layer is slope.
  • Select by attribute from zonal statistics table. Slope >=20. Create layer from selection.

Accounting for light exposure:

Identifying potential problem buildings:

       Create 80ft buffer around All Vacant.

       From buildings layer, select by BLDG_HGT, height >= 40ft -> create layer.

       Select by location: buildings_40ft; Source layer: buildings_buffer -> intersects target>create layer.

Tree density (5 trees per 10,000 square feet)

       Street trees only.

  • Select by attribute street trees only -> create new layer.


       Number of trees per parcel

  • Spatial join -> Join by special attribute: Trees to Parcels
  • Add field: tree_dens -> calculate field: count of trees/land_sf.
  • Select out those with tree_dens >= .0005.


Defining School PedSheds:

A PedShed is defined as an area that is within walking distance of some facilities. In this case, the facilities are Boston Public Schools and the walking distance is 500 meters.

To create PedSheds:

       Export roads in class 3-6 to their own shape file.

       Network Analyst—Service Area Analysis

       Facilities—Load Locations (Schools)

       Service Area Properties— Create a ¼ and ½ Mile Walkable Zones layer. [PBM1]

  • Follow specifications in Proximity Tools tutorial.

       Create new shape files of both the polygons and lines.


Close Up on Grove Hall:

       Created an outline of grove hall through using the “select by lasso tool” on the BostonNeighborhoods00 layer, created layer from selection.

  • Dissolved neighborhoods using: Toolbox: Data Management Tools: Generalization: Dissolve: ID



Annotated Bibliography:

Chin, D. (2013). Introduction to GIS (Final Paper). 7May

Denise Chin’s project helped me to think through the parameters for an urban farm. By using the same parameters that she used with her team, I can maintain consistency in Boston Urban Agriculture tools. Additionally, by using the same process that Chin used to determine eligible plots of land, I can then add another component to the map in order to map sites that could be prioritized for educational purposes.

GIS Best Practices: US Department of Agriculture. August 2006. ( )

This website is helping me to understand how farmers might use the visuals to inform their operations. Although the community is the primary audience for this map, the hope is that the prioritized land with eventually be used by farmers. If I can include useful information on the maps, it further reduces barriers to establishment of new farms.

Brown, Katherine H. et al. Urban Agriculture and Community Food Security in the United States: Farming from the City Center to the Urban Fringe . Urban Agriculture Committee of the Community Food Security Coalition. February 2002.

This paper contextualizes Urban Agriculture in terms of food security and food justice. This is especially useful for the Grove Hall Initiative, as this neighborhood is comprised of a primarily minority, low-income population. By understanding how Urban Agriculture can create a more just food system, I can better prioritize information that would be useful in a map for a neighborhood development project. Although Grove Hall is where the current project is focused, other neighborhoods in Boston could also benefit from a map with a justice orientation.


Thapa, Rajesh Bahadur and Yuji Murayama. “Land evaluation for peri-urban agriculture using analytical hierarchical process and geographic information system techniques: A case study of Hanoi.” Land Use Policy, Volume 25, Issue 2, Pages 225-239: April 2008. Accessed from Science Direct, November 6, 2008.


This paper has helped me understand suitability criteria for urban farms to confirm Chin’s criteria. Additionally, there is the potential to use their categorization criteria to evaluate high/medium/low suitability levels or each piece of land.



[i] Ratcliffe, M. M., Merrigan, K. A., Rogers, B. L., & Goldberg, J. P. (2011). The effects of school garden experiences on middle school-aged students’ knowledge, attitudes, and behaviors associated with vegetable consumption. Health Promotion Practice ,   12 (1), 36-43.


[ii] Lineberger, S. E., & Zajicek, J. M. (2000). School gardens: can a hands-on teaching tool affect students' attitudes and behaviors regarding fruit and vegetables?.   HortTechnology ,   10 (3), 593-597.


[iii] Heim, S., Stang, J., & Ireland, M. (2009). A garden pilot project enhances fruit and vegetable consumption among children.   Journal of the American Dietetic Association ,   109 (7), 1220-1226.


[PBM1] Just note that having two zones makes things a bit more complex.