Tree Canopy UTC_Report_Burlington
1/4/2008 1 Tree Canopy 39% (43%) Low Lying Vegetation, 18% (20%) Pavement or Bare Soil, 20% (22%) Agriculture, 5% (5%) Wetland, 6%Water, 4% Structures, 8% (9%) Imagery NLCD Canopy Summarry
proach based on high resolution
imagery found that 2648 acres of Burlington is covered by tree canopy (termed Existing UTC). This corresponds to 39% of the �
� not occupied by water or wetland). An additional 36% (2198 acres) of Burlington could conceivably
be covered by urban tree canopy (termed Possible UTC).
located in areas of residential land use. use. Residential land also contains most of the Possible UTC (34% of the land area, 745 acres). UTC enhancement can be most
efficiently realized by maximizing protection and maintenance in combination with new plantings and natural regeneration. The impacts of setting a UTC goal will likely include focusing
or reallocating public agency resources (funds, staff, etc.) to enhance UTC urban open land. On private lands, a combination of education and outreach, landowner and redevelopment incentives,
and refocusing of regulatory mechanisms to specifically achieve the objectives of the UTC goal will likely be required. Prrojjectt Backgrround �
� carried out by the Spatial Analysis Laboratory (SAL) of the
Northern Research Station
at the request of the City of Burlington. The project was completed in collaboration with the Vermont Department of Forest, Parks and Recrea-
The goal of the project was to apply the USDA Forest Ser-
The UTC assessment protocols make use of high resolution geospatial datasets, enabling UTC metrics to be computed at the parcel level. UTC metrics provide
detailed informa- UTC goal setting. This project sought to leverage
existing investments in geospatial data made by the city, enabling the analysis to be completed with minimal cost.
and Possible Urban Tree Canopy Devellopmentt off a Hiigh Ressolluttiion Land Coverr Dattassett The need for high resolution land cover Land cover datasets lack both the accuracy
and the resolution to effectively map tree canopy in urban areas. The
very valuable for regional analysis but with a relatively coarse resolution (30 meters) it fails to capture all of the tree canopy in the urban forest (Figure 1). Capitalizing on existing
data investments In 2004 the City of Burlington participated in the Chitten-
purchase of high resolution imagery and high resolution elevation data (know as LIDAR). Leveraging this existing investment in high quality data automated feature extraction technology,
a sub-meter, seven class land cover map was created at a fraction of what it would cost to map it manually (Figure 2). This detailed assessment enabled the compilation of city-wide land
cover (Figure 3) and parcel based UTC metrics. Imagery & LIDAR Land Cover Figure 1: NLCD 2001 Canopy comparison Figure 2: Automated land cover mapping steps Figure 3: Land cover summary.
Percentages are based on the city area using the
percent by land area
1/4/2008 2 % Land Area % Land Use % UTC Type % Land Area % Land Use % UTC Type % Land Area % Land Use % UTC Type Agriculture 1% 19% 2% 0% 3% 0% 0% 1% 0% Commercial 2% 27% 6% 2% 22% 10%
2% 27% 16% Industrial 0% 16% 1% 0% 22% 1% 0% 37% 3% Mixed Com/Res 0% 27% 0% 0% 7% 0% 0% 28% 1% Residential 16% 49% 39% 7% 21% 36% 4% 14% 31% Tax-Exempt 7% 42% 17% 3% 22% 19% 2% 14% 15%
Unknown 3% 53% 7% 1% 26% 7% 1% 13% 5% Urban Open Land 10% 36% 26% 4% 15% 24% 4% 12% 25% Utility 1% 24% 2% 0% 8% 2% 0% 12% 3% Existing UTC LandUse Possible UTC (Vegetation) Possible UTC
(Impervious/Bare Soil) Exiissttiing and Possssiiblle UTC Parrcell & Land Usse Summarry UTC metrics for the City of Burlington were computed using the UTC assessment protocols. The UTC
protocols integrate the land Existing UTC was computed
by simply summarizing the tree canopy land cover class. Two types of Possible UTC were computed: Possible UTC (Vegetation) and Possible UTC (Impervious/Bare Soil). Possible UTC (Vegetation)
was computed by finding all areas in the (Vegetation)
excludes all wetland and agricultural vegetation. Possible UTC (Impervious/Bare Soil) was computed by summariz-
the roadways. It is typically easier to increase tree canopy on Possible UTC (Vegetation) as compared to Possible UTC (Impervious/Bare
Soil) areas. Parcel Boundaries Existing UTC Possible UTC % Land Use = Area of UTC type for specified land use Area of all land for specified land use The % Land Use value of 49% for
Existing UTC residential land indicates that 49% of residential land is covered by tree canopy. % UTC Type = Area of UTC type for specified land use Area of all UTC type type The % UTC
Type value of 39% for Existing UTC residential land indicates that 39% of Existing UTC lies in residential land use. % Land Area = Area of UTC type for specified land use Area of all
land The % Land Area value of 16% for Existing UTC residential (excluding
water and wetland) is residential tree canopy. Following the computation of the Existing and Possible UTC the UTC �
� base (Figure 4). For each parcel the absolute Existing and Possible UTC was computed along with the percent of Existing UTC and Possible
UTC (UTC /area of the parcel). Using the land use information associated with each parcel UTC metrics were summarized by land use category (Figure 5). Table 1 shows how for each land
use category UTC metrics were computed as a percent of all land within the particular land use category (% Land Use), as a percent of the UTC type (% UTC Type) and as a percent of the
area of all land (% Land Area). % Land Use allows for comparison of Existing UTC and Possible UTC in a given land use class while % UTC Type allows the relative contribution of a land
use category to either the Existing or Possible UTC. Figure 4: Parcel-based UTC metrics Figure 5: UTC metrics summarized by land use 0 500 1000 1500 2000 2500 Agriculture Commercial
Industrial Mixed Com/Res Residential Tax-Exempt Unknown Urban Open Land Utility Area (acres) Existing UTC Possible UTC (Vegetation) Possible UTC (Impervious/Bare Soil) Not Suitable Table
1: UTC metrics by type, summarized by land use
1/4/2008 3 Northeastern States Research Cooperative David White City Planner City of Burlington Danielle Fitzko U&CF State Coordinator VT Dept. of Forest, Parks, & Recreation Agency
of Natural Resources Morgan Grove Research Scientist Northern Research Station USDA Forest Service mgrove@fs.fed.us 802.951.6771 x1111 -Dunne Geospatial Analyst
Spatial Analysis Laboratory Rubenstein School of the Environment & Natural Resources University of Vermont joneildu@uvm.edu 802.656.3324 Prepared for: Prepared by: Funded by: Trees for
Local Communities grant from the Department of Forests, Parks and Rec-reation's Urban and Community Forestry Program. USDA Forest Service State & Private For-estry �
�land) is covered by tree canopy (Existing UTC) that encom-passes 2648 acres. 36% (2198 acres) of the land area
in the city is not a road a structure or being used for agriculture, and thus could could conceivably support tree canopy. This Possible UTC is almost evenly split between impervious
surfaces/bare soil and low-lying vegetation. The majority of land in Burlington (32%) falls into the resi-dential category. Residential land has a relatively high percentage of Existing
UTC, with 49% of all residential land covered by tree canopy. 36% of residential land can be classified as Possible UTC.
open land is primarily composed
of the rights-of-way along roads, some parks, institutional land (e.g. UVM, school grounds), and open space (e.g. Starr Farm). Al-though only 36% of urban open land is Existing UTC,
only 27% of the urban open land can be considered to be Pos-sible UTC. Within the ROW, 20% is Existing UTC and 1% is Possible UTC (Possible UTC does not account for the UTC that could
overhang a road). Commercial and industrial land have noticeably low amounts of Existing UTC (27% and 16% respectively) and high proportions of Possible UTC (49% and 59%) respec-tively.
Unlike residential land where the majority of Possi-ble UTC is low-lying vegetation, on commercial and indus-trial land it is impervious surfaces (Figure 7). ResultsResults ConclusionsConclusions
Investments in remotely sensed data such as the imagery and LIDAR acquired as part of the CCMPO purchase in 2004 provide a robust �
�
percentage points. This report presents broad generalizations. With Existing UTC and Possible UTC summarized at the parcel level and integrated with the
examined and targeted for UTC improvement.
�
�wildlife. Burlington has room to increase its UTC through a combina-tion of maintenance, tree plantings, and natural regeneration. Although residential
land appears to be built-out with respect to tree canopy, analysis of individual parcels shows that there are a number of properties that have very low amounts of Existing UTC. Incentives
or educational initiatives could be employed to encourage these members of the community to increase UTC on their properties. �
�ued due to the numerous benefits they afford, but street tree plant-ings alone will not be able to substantially increase
the UTC in Bur-lington. Large parcels of land with high Possible UTC that are owned by the government or institutions will likely offer the best opportuni-ties for UTC enhancement. The
relatively low amount of Existing UTC combined with the high Possible UTC in commercial and industrial areas indicates that there is a considerable need to green these land use types.
Increasing the UTC on industrial and commercial land could help to fragment the connected impervious surfaces and improve water quality and aes-thetics. This is particularly important
given the proximity of some these highly impervious parcels to Lake Champlain. Figure 6: Urban open land with a large amount of Possible UTC in the low lying vegetation category. Figure
7: Industrial land with a large amount of Possible UTC in the imper-vious category