F.2.a

Tree Canopy Coverage & Change Detection City of Brookhaven, Georgia | 2010 – 2015 Final Results

2015 Natural Color (left) and Infrared Color (right) Imagery of Peachtree Road and Dresden Drive in Brookhaven, Georgia

PREPARED BY MIKE EDELSON, GIS MANAGER, INTERDEV JANUARY 2017

City of Brookhaven 4362 Peachtree Rd Brookhaven, GA 30319 404.637.0500 http://www.brookhavenga.gov/

InterDev - Atlanta 2650 Holcomb Bridge Road Suite 310, Alpharetta, GA 30022 770.643.4400 http://www.interdev.com/

Packet Pg. 11

Attachment: 2016_Brookhaven_TreeStudyResults (2601 : 2016 Tree Canopy Study Presentation - Mike Edelson)

2016

Tree Canopy Coverage & Change Detection: Brookhaven, Georgia 2016 Report Overview and Background

The City of Brookhaven, Georgia commissioned a study of the urban tree canopy. The purpose of the study is to assist the City by providing key decision makers and citizens a quantitative method for monitoring tree canopies. Performing this study over the course of several years is a powerful tool for understanding and monitoring important trends in your tree canopy. In 2014, the City of Brookhaven started the process of revising the City’s tree ordinance. The Community Development Director, City Arborist, and InterDev GIS proactively developed an innovative method to accurately measure the City’s tree canopy. The study’s final results showed the canopy coverage for 2010 (51.7%) and 2013 (49.0 %.) More importantly this study established a baseline measurement for monitoring Brookhaven’s tree canopy into the future. Using satellite imagery obtained through U.S. Department of Agriculture’s National Agricultural Imagery Program (NAIP), this study performs a series of analyses using advanced GIS tools and processes to extract the tree canopy within the city limits. The NAIP imagery Location Map program was developed by the U.S. Department of Agriculture to provide satellite imagery during the “leaf‐on” season that is used for a variety of different applications, mostly in the farming, agricultural, or timber industries. This study uses this imagery in the urban setting to help calculate Brookhaven’s tree canopy. This study does not aim to highlight the numerous important factors of maintaining the tree canopy nor does this study aim to provide recommendations for changes to ordinances or policies that are used for tree preservation and replacement. This analysis aims to provide the city with a measurement of tree canopy for each square meter of the City and overall percent coverage. During the past 3 years, Brookhaven has seen unprecedented development which caused citizens to have legitimate concerns over the valuable tree canopy. By commissioning this study, Brookhaven will continue to develop an authoritative method to continuously monitor the change in tree canopy. Our hope is that this information will facilitate informed and educated conversations around preserving Brookhaven’s valuable tree canopy.

InterDev | Atlanta – Chicago

1

Packet Pg. 12

Attachment: 2016_Brookhaven_TreeStudyResults (2601 : 2016 Tree Canopy Study Presentation - Mike Edelson)

F.2.a

Brookhaven, GA Tree Canopy Study December 2016

F.2.a

Brookhaven, GA Tree Canopy Study December 2016

Findings Percent Coverage of Canopy Total City-Wide Consolidated Zoning Category Single Family Residential Areas Multifamily Residential Areas Commercial/Industrial Areas

2010 51.7%

2013 49.0%

2015 53.6%

61.8% 42.7% 37.3%

59.6% 46.7% 37.7%

60.7% 49.7% 38.8%

Final results show a small gain in total tree canopy over the past 5 years. For the purpose of this study, our goal was to measure the City’s canopy by total area and smaller areas of consolidated zoning categories. This allows decision makers to further dive deeper into the results and explore in finer detail where canopy was either lost or gained. The City’s zoning districts were consolidated into 3 reporting areas: single family residential, multifamily residential and commercial/industrial. The areas highlighted in yellow are identified as single family residential, the areas highlighted in purple are multiple family residential, and the areas highlighted in red are identified as commercial or industrial (see map on right). Although our findings have identified several localized areas of major canopy loss of 5 acres or more (see maps on pages 6 and 7), overall the City’s tree canopy was relatively unchanged or slightly increased. Based on our analysis, the City has an estimated 3,928 acres of tree canopy out of a total of 7,420 acres. Studies such as this and others have indicated that over the course of a few years the slow, normal growth of a tree canopy of this size has been enough to offset the rate of trees being lost on single lots and a few large developments. Age, condition, and the environmental factors associated with urban forest management is an extremely important factor and should be considered when using this data to analyze the health of your urban forest. Determining and analyzing these factors are outside the scope of this study. The following pages contain a series of maps identifying tree canopy coverage and tree canopy change between the years of 2010 and 2015.

InterDev | Atlanta – Chicago

2

Packet Pg. 13

Attachment: 2016_Brookhaven_TreeStudyResults (2601 : 2016 Tree Canopy Study Presentation - Mike Edelson)

The following table is an overview of our findings:

Attachment: 2016_Brookhaven_TreeStudyResults (2601 : 2016 Tree Canopy Study Presentation - Mike Edelson)

Brookhaven, GA Tree Canopy Study December 2016

InterDev | Atlanta – Chicago

F.2.a

3

Packet Pg. 14

Attachment: 2016_Brookhaven_TreeStudyResults (2601 : 2016 Tree Canopy Study Presentation - Mike Edelson)

Brookhaven, GA Tree Canopy Study December 2016

InterDev | Atlanta – Chicago

F.2.a

4

Packet Pg. 15

Attachment: 2016_Brookhaven_TreeStudyResults (2601 : 2016 Tree Canopy Study Presentation - Mike Edelson)

Brookhaven, GA Tree Canopy Study December 2016

InterDev | Atlanta – Chicago

F.2.a

5

Packet Pg. 16

Attachment: 2016_Brookhaven_TreeStudyResults (2601 : 2016 Tree Canopy Study Presentation - Mike Edelson)

Brookhaven, GA Tree Canopy Study December 2016

InterDev | Atlanta – Chicago

F.2.a

6

Packet Pg. 17

Attachment: 2016_Brookhaven_TreeStudyResults (2601 : 2016 Tree Canopy Study Presentation - Mike Edelson)

Brookhaven, GA Tree Canopy Study December 2016

InterDev | Atlanta – Chicago

F.2.a

7

Packet Pg. 18

F.2.a

Brookhaven, GA Tree Canopy Study December 2016

Methodology

Imagery Imagery was obtained through the U.S. Department of Agriculture National Agricultural Imagery Program (USDA NAIP). The NAIP imagery program provides 1-meter resolution aerial imagery collected during the “leaf-on” season on a 3-5-year cycle. NAIP imagery is important because it attempts to produce near identical conditions in imagery collection over the span of 3-5 years. A critical factor in the determination of tree canopy coverage over time is the ability to compare two aerial imagery data sets taken years apart that are as close to identical as possible. Identical conditions include variables such as time of day, time of year, pixel size, camera angle, and cloud cover as well as other variables. Below is side-by-side aerial imagery showing tennis courts and a swimming pool in 2009 and 2015. Imagery on the left was collected in July 2009. Imagery on the right was collected in July 2015. Notice the shadows over the pool and tennis courts, they are almost in the same location indicating these 2 aerial photos were taken at approximately the same time of day but 6 years apart.

2009 NAIP Imagery

InterDev | Atlanta – Chicago

2015 NAIP Imagery

8

Packet Pg. 19

Attachment: 2016_Brookhaven_TreeStudyResults (2601 : 2016 Tree Canopy Study Presentation - Mike Edelson)

Aerial imagery collected in 2010, 2013 and 2015 is analyzed to measure the canopy coverage and change over time.

F.2.a

Brookhaven, GA Tree Canopy Study December 2016

The processing technique used in this study is called the Iso Cluster Unsupervised Classification tool. This process analyzes each pixel in the image to find natural clusters of values (or colors.) It then programmatically combines millions of pixels values down to 100 different values based on similar pixel value characteristics. In essence it automatically simplifies the colors of the image. Once all the pixel values are combined into 100 manageable values, our analyst reviews each of the 100 values and classifies them as Canopy, Non-Canopy, or Unknown. Basically, we are training the database: “We’ve identified this pixel as being a tree canopy, now go ahead and identify all similar pixel values as tree canopy.” This command then programmatically assigns all other pixels that have similar values as a tree canopy. This process is then repeated 100 times (one for each of the 100 Iso Cluster Unsupervised Classification classes). If a value is classified as Unknown, our analyst will then repeat the Iso Cluster Unsupervised Classification process again only on the Unknown values and the process is repeated until there are no more Unknowns. This study focuses on high canopy only, as shown in the graphic below. Small shrubs and low lying vegetation were not considered as part of the tree canopy. In some cases, very thick low lying vegetation (as in abandoned empty lots) will show up as tree canopy. Our study attempts, through our Quality Control (QC) process (see next section) to limit this vegetation from being classified as canopy cover. For the purpose of this study we are classifying all low vegetation as Non-Canopy.

This image shows examples of vegetation that are classified as Tree Canopy. Image Source: Arbor Day Foundation

InterDev | Atlanta – Chicago

9

Packet Pg. 20

Attachment: 2016_Brookhaven_TreeStudyResults (2601 : 2016 Tree Canopy Study Presentation - Mike Edelson)

Analysis

F.2.a

Brookhaven, GA Tree Canopy Study December 2016

Once our analysis is complete, our QC Lead visually inspected and identified major errors in the classification process. Errors are identified and corrected in the method outlined below. In this example, the baseball fields at Blackburn Park have been identified as having major errors. 1. 2. 3. 4.

Several thick patches of tree shadows are incorrectly classified as tree canopy on the baseball field. Our analyst created a polygon-shaped baseball field over the area that needs to be corrected Pixels are reclassified within the baseball field layer to Non-Canopy All pixels on the baseball field are classified correctly

InterDev | Atlanta – Chicago

10

Packet Pg. 21

Attachment: 2016_Brookhaven_TreeStudyResults (2601 : 2016 Tree Canopy Study Presentation - Mike Edelson)

Quality Control (QC)

F.2.a

Brookhaven, GA Tree Canopy Study December 2016

Using a random point generator tool, we generated 100 points within the City limits. Each one of those points were manually classified by our analyst as Canopy or Non-canopy. We compared those classifications with the classifications determined in the automated classification process described above. To pass our accuracy threshold we required a score of 90% or higher. A review of several other tree canopy studies indicated that an accuracy assessment greater than 90% is well above the industry standard for this type of study. If less than 90 points matched, then our team would go back and fix the errors until we received a higher than 90% accuracy threshold. The graphic below shows this process:

Red dots indicate the location of random samples used to determine accuracy

Flowchart describing the overview of the analysis, QC, and accuracy assessment

InterDev | Atlanta – Chicago

11

Packet Pg. 22

Attachment: 2016_Brookhaven_TreeStudyResults (2601 : 2016 Tree Canopy Study Presentation - Mike Edelson)

Determining Accuracy Threshold

F.2.a

Brookhaven, GA Tree Canopy Study December 2016

Comparison Cities

Atlanta Region Comparison Cities City of Decatur City of Atlanta City of Avondale Estates City of Sandy Springs City of Brookhaven

Year 2010 2014 2015 2015 2015

Percent Tree Coverage 45.1% 47.9% 54.0% 62.0% 53.6%

This table was developed and published in the 2014 City of Atlanta Urban Tree Canopy Study by the Georgia Tech Center for Geographic Information Systems. This table compares urban tree canopy percentages (UTC %) with other major metropolitan areas in the United States.

Percent Tree Canopy Estimates. Source: Atlanta Urban Tree Canopy

Note •



Due to a large annexation that occurred on the south end of the city in January of 2015, some discrepancies may be embedded in the results. When this occurs, this study does it’s best to either note the discrepancies or standardize the results to best explain the change in canopy. This study does not include the recent Woodcliff Drive annexation.

InterDev | Atlanta – Chicago

12

Packet Pg. 23

Attachment: 2016_Brookhaven_TreeStudyResults (2601 : 2016 Tree Canopy Study Presentation - Mike Edelson)

A quick comparison of tree canopy coverage for other cities in the Atlanta area are shown below.

F.2.a

Brookhaven, GA Tree Canopy Study December 2016

National Agriculture Imagery Program (NAIP) Aerial Imagery. Salt Lake City, UT: United States Department of Agriculture Farm Service Agency, 2015. http://www.fsa.usda.gov/programs-and-services/aerialphotography/imagery-programs/naip-imagery/ City of Brookhaven Tree Protection and Replacement Ordinance. Brookhaven, GA. Available: http://www.brookhavenga.gov/home/showdocument?id=1553 Georgia Cities. Atlanta, GA: Atlanta Regional Commission Open Data Portal, 2016. Available: http://arc.garc.opendata.arcgis.com/ Brookhaven GIS City Limits and Zoning Categories. Brookhaven, GA: Brookhaven WebApp GIS REST Service Endpoint Page, 2016. Available: http://gis.brookhavenga.gov/arcgis/rest/services/ Assessing Urban Tree Canopy in the City of Atlanta. Atlanta, GA. City of Atlanta and Georgia Institute of Technology, 2014. Available: http://www.prism.gatech.edu/~ag124/UTCFinalReport.pdf Tree Canopy Study. Sarasota County, FL. Sarasota County Environmental Protection Division, 2013. Available: https://www.scgov.net/trees/Documents/2013%20Tree%20Canopy%20Study.pdf Classification Accuracy Assessment. State College, PA. Pennsylvania State University. Available: https://www.e-education.psu.edu/geog883/node/524 Tree Height Diagram. Arbor Day Foundation. Available: https://www.arborday.org/programs/treecityusa/bulletins/images/004-figure-tree-height-diagram.gif City of Decatur Tree Canopy Conservation Ordinance. Decatur, GA. Available: http://www.decaturga.com/home/showdocument?id=5196 The Future of Large Old Trees in Urban Landscapes. Le Roux DS, Ikin K, Lindenmayer DB, Manning AD, Gibbons P. Webb E, ed. PLoS ONE. 2014;9(6):e99403. doi:10.1371/journal.pone.0099403. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4062419/

InterDev | Atlanta – Chicago

13

Packet Pg. 24

Attachment: 2016_Brookhaven_TreeStudyResults (2601 : 2016 Tree Canopy Study Presentation - Mike Edelson)

References and Data Sources

Tree Canopy Study.pdf

Final Results. F.2.a. Packet Pg. 11. Attachment: 2016_Brookhaven_TreeStudyResults (2601 : 2016 Tree Canopy Study Presentation - Mike Edelson). Page 1 of ...

3MB Sizes 0 Downloads 126 Views

Recommend Documents

relationship among tree canopy quantity, community ...
The Tree City USA (TCUSA) program is a cooperative .... A multivariate stepwise regression analysis was run on ..... techniques to land use classification.

LiDAR- Based Estimation of Canopy Fuel Parameters of Tree ...
the fieldwork and LiDAR was done. R2 was also computed to test relationship. between the two datasets. VILLAR, R. G. et al. - CMUJS Vol. 20, No.3 (2016) 140-149. Page 3 of 10. LiDAR- Based Estimation of Canopy Fuel Parameters of Tree Plantation in Bu

Mapping Tropical Rainforest Canopy Disturbances ... - Semantic Scholar
4 Feb 2013 - COSMO-SkyMed scenes, the coherence images could not be used as meaningful additional information to detect forest disturbances in our study. The processing chain for combined InSAR Stereo DSM extraction is illustrated in Figure 3. Co-reg

(Gossypium hirsutum) to foliar nutrition and canopy management ...
canopy management practices. Madras Agric. J. 91 (7-12) : 370-373 ... may reduce yield and quality of lint due to early drought or excess rainfall in different.

Estimation of the Minimum Canopy Resistance for Croplands and ...
Nov 4, 2008 - Science, 309, 570–574. Fulton, R. A., J. P. Breidenbach, D.-J. Seo, D. A. Miller, and T. O'Bannon, 1998: The WSR-88D rainfall algorithm. Wea.

Estimation of the Minimum Canopy Resistance for ...
Nov 4, 2008 - cloud cover and aerosol content, the fraction of K↓ constituting PAR ...... T. Holt, S. Zhong, P. C. Pyle, and J. Basara, 2006: Urban and. 4468.

Nectarine tree named 'CAKEREDAL'
Jan 29, 2013 - No damages Were caused by ascer tained temperatures as loW as —120 degrees Celsius. The tree Was also very resistant to frosty springtime.

Apple tree named 'Candy'
(57). ABSTRACT. A new and distinct apple tree is disclosed. 'Candy,' a limb. (21) Appl. No.: 11/400,782 sport of 'Aztec,' is notable for its distinctive fruit, which. _.

Plum tree named 'Suplumfortyone'
Jan 19, 2010 - Fruit use: Fresh market. 0 Fruit shipping and keeping quality: Good, holds ... Width of Stalk End: Medium; approximately 3 mm. Angle of Stalk ...

Avocado tree named 'Maluma'
Oct 28, 2008 - *Grooves and ridges. Bark color.%ireyed-green 194B. Trunk lenticels.4Color greyed-green 194A, siZe 3 mm by 6 mm, density 150 per 100 cm2 ...

Apricot tree named 'ASFCOT0404'
Mar 9, 2009 - (12) United States Plant Patent (10) Patent N0.: US PP21,138 P3 ... life after harvesting, in order to facilitate long-distance ship ping. Our ?nal ...

Apple tree named 'Jugala'
'Jugala' is a Gala-type apple characterized by its early harvest. 3 date as compared to other known Gala varieties. (51) Int. Cl. A01H 5/00. (2006.01). 5 Drawing ...

Estimation of the Minimum Canopy Resistance for ... - Semantic Scholar
Nov 4, 2008 - ence of air temperature (F4) assumes all vegetation photosynthesizes ...... vegetation network during the International H2O Project. 2002 field ...

Apple tree named 'Jugala'
'Jugala' is a Gala-type apple characterized by its early harvest. 3 date as compared to other known Gala varieties. (51) Int. Cl. A01H 5/00. (2006.01). 5 Drawing ...

Family Tree Maker - MOBILPASAR.COM
Sep 21, 2006 - ... MOLINA Y MÁRQUEZ and LORENZA RUIZ DE ESPARZA. She was born. 1625 in Aguascalientes, Ags. Mexico, and died Aft. 1693. More About FRANCISCA DE TISCAREÑO AKA FRANCISCA GABAI: Baptism: 24 Mar 1625, El Sagrario, Aguascalientes, Ags. M

Nectarine tree named 'NECTARPERF'
Jan 29, 2013 - Foreign Application Priority Data. Latin pIaIneZ ... nated 'NECTARPERF', has a large fruit of very long shelf a life without ... The tree is of large siZe and is vigorous. Fruit can .... Color of mature bI'III'IChKSIiBI'OWII (RHS GREY.

Pecan tree named 'Morrill'
Jul 8, 2011 - The colors of an illustration of this type may vary with lighting and other .... (cc) (g) (mm) widthZ ratioy (mm) (%) date'C. 'Byrd'. 11.3a 7.8a 42.5a ... end of the peanut harvesting season, making it a more suitable early cultivar for

Peach tree named 'Flatbeauti'
Nov 20, 2012 - (12) United States Plant Patent (10) Patent No.: US PP23 ... nated 'FLATBEAUTI', has a large ?at fruit of very long shelf ..... distance shipping.

Nectarine tree (Summer Bright)
Oct 24, 1989 - color code designations are by reference to the Inter. Society Color Council .... Market: Fresh and long distance shipping. Keeping quality: ...

Cherry tree named 'RR2a'
Jul 2, 2008 - (45) Date of Patent: ... maturity date (approximately 36 days later), ?rmness (ap ... tain its ?rmness for up tWo months in cold storage.

Recall Tree - GitHub
Dec 10, 2016 - but how to define close? Computationally expensive. ▻ . . . attempt to build an index. Competitive when all classes have singleton support.

Tree Removal Rancho Cucamonga Links - (909) 963-1425 - Tree ...
Tree Removal Rancho Cucamonga Links - (909) 963-1425 - Tree Removal Rancho Cucamonga.pdf. Tree Removal Rancho Cucamonga Links - (909) ...