|
Arboriculture & Urban Forestry 34(6): November 2008 387
<br />to illustrate how plot size affects the total tree and standard error
<br />estimate. Average plot time for field plot setup, cover estimates,
<br />and measurements per tree were used to estimate how average
<br />field measurement time would likely vary as tree cover changes.
<br />In a separate analysis, an additional test of plot size and plot
<br />design was conducted using GIS tree cover, land use, and parcel
<br />data for the city of Syracuse. Five hundred points were randomly
<br />distributed throughout the city. At each point, the following
<br />seven different plot sizes or designs were constructed around the
<br />point using GIS: 1) one -twenty-fourth acre (0.017 ha) circular
<br />plot; 2) one -twelfth acre (0.034 ha) circular plot; 3) one -tenth
<br />(0.04 ha) circular plot; 4) one -eight acre (0.05 ha) circular plot;
<br />5) one -sixth acre (0.067 ha) circular plot; 6) one-fourth acre (0.1
<br />ha) circular plot; and 7) four one -twenty-fourth acre (0.017 ha)
<br />circular plots (cluster plot) using the USDA Forest Service For-
<br />est Inventory and Analysis (FIA) plot design (USDA Forest
<br />Service 2000). With this cluster plot design, three subplots were
<br />established 120 ft (36.6 m) from the center subplot at 120°, 240°,
<br />and 360° azimuths.
<br />For each of the plot sizes and designs, total amount of tree
<br />cover within the plot was assessed using a 2 ft (0.61 m) resolu-
<br />tion tree cover map (Myeong et al. 2003), and the number of
<br />parcels and associated number and area of land uses in each
<br />parcel within the plot design was recorded using a digital land
<br />use parcel map. The average amount of permissions required for
<br />each plot design was categorized among three classes: 1) per-
<br />mission required (residential land use parcels); 2) permission
<br />questionable -uncertain if crew would need to obtain permis-
<br />sion (commercial/industrial, institutional, utility/transportation
<br />parcels); and 3) no permission needed (greenspace, street right-
<br />of-ways, and vacant parcels) to assess how permissions would
<br />vary based on plot size and design. The average percent of plot
<br />area within the parcel that contained the plot center was also
<br />calculated. This calculation was done to help determine how
<br />much of the plot area would require the crew to move to an
<br />additional parcel and how much of that extra plot space would
<br />require additional permissions. Mean tree cover and standard
<br />error for each plot design were calculated and compared with the
<br />actual tree cover as classified by the tree cover map.
<br />Effect of Sample Size on Total Population
<br />Estimate Precision
<br />To determine the effect of sample size on the standard error
<br />estimate for the total tree population, sample data from 14 cities
<br />were analyzed using the UFORE model (Nowak and Crane
<br />2000; Nowak et al. 2002) (Table 1). For each city, population
<br />total, standard error (SE), and relative SE were calculated. The
<br />relative SE is a measure of estimated reliability and is the ratio
<br />of SE to the estimate, in this case, population total (SE/total x
<br />100) (US Department of Health and Human Services, Centers for
<br />Disease Control and Prevention 2007). Eleven of the cities were
<br />sampled using a stratified random sampling approach, and three
<br />using a randomized grid approach, which was used to facilitate
<br />long-term monitoring of urban forest change. Standard error for
<br />each city was standardized to a population size of 200 plots using
<br />the formula: SE = standard deviation/Ain. The average SE using
<br />200 plots was calculated for the 14 cities and used to illustrate
<br />how SE of the total tree population estimate will vary as sample
<br />size varies between 10 and 500 plots.
<br />RESULTS
<br />Effect of Plot Size on Data Collection Time and
<br />Total Population Estimate Precision
<br />Increasing plot size from a one -twenty-fourth acre (0.017 ha)
<br />plot to a one -sixth acre (0.067 ha) plot nearly doubled the
<br />amount of time needed to measure the plot variables, but also
<br />nearly cut in half the relative standard error for the total popu-
<br />Table 1. Estimates of total number of trees and standard errors from 14 cities analyzed using the UFORE model.z
<br />Number of trees 200 plot`"
<br />City Total SE Year No. plots
<br />Atlanta, GA'
<br />Baltimore, MD"
<br />Boston, MA'
<br />Freehold, NJ"
<br />Jersey City, NJ"
<br />Minneapolis, MN'
<br />Moorestown, NJ"
<br />Morgantown, WV`
<br />New York, NYw
<br />Philadelphia, PA'
<br />San Francisco, CAr
<br />Syracuse, NY"
<br />Washington DC"
<br />Woodbridge, NJ"
<br />9,415,000
<br />2,571,000
<br />1,183,000
<br />48,000
<br />136,000
<br />979,000
<br />583,000
<br />658,000
<br />5,212,000
<br />2,113,000
<br />668,000
<br />876,000
<br />1,928,000
<br />986,000
<br />749,000
<br />494,000
<br />109,000
<br />6,000
<br />22,000
<br />165,000
<br />53,000
<br />79,000
<br />719,000
<br />211,000
<br />98,000
<br />119,000
<br />224,000
<br />97,000
<br />1997
<br />2004
<br />1996
<br />1998
<br />1998
<br />2004
<br />2000
<br />2004
<br />1996
<br />1996
<br />2004
<br />2001
<br />2004
<br />2000
<br />205
<br />200
<br />217
<br />144
<br />220
<br />110
<br />206
<br />136
<br />206
<br />210
<br />194
<br />197
<br />201
<br />215
<br />SE RSE
<br />758,000
<br />494,000
<br />114,000
<br />5,000
<br />23,000
<br />122,000
<br />54,000
<br />65,000
<br />729,000
<br />216,000
<br />97,000
<br />119,000
<br />224,000
<br />100,000
<br />Average relative standard error = 12.1%.
<br />'Estimated standard error (SE) and relative standard error (SE/total x 100; RSE) using a sample of 200 one -tenth acre (0.04 ha) plots.
<br />"Str. random = stratified random sample; random grid = randomized grid sample.
<br />'Data collection by ACRT, Inc.
<br />"Data collection by U.S. Forest Service.
<br />"Data collection by New Jersey Department of Environmental Protection.
<br />`Data collection by Davey Resource Group.
<br />'Data collection by West Virginia University.
<br />'Data collection by city personnel.
<br />"Data collection by Casey Trees Endowment Fund.
<br />8.1
<br />19.2
<br />9.6
<br />10.1
<br />16.7
<br />12.5
<br />9.3
<br />9.9
<br />14.0
<br />10.2
<br />14.5
<br />13.5
<br />11.6
<br />10.2
<br />Sample'
<br />Str. random
<br />Str. random
<br />Str. random
<br />Str. random
<br />Str. random
<br />Random grid
<br />Str. random
<br />Str. random
<br />Str. random
<br />Str. random
<br />Random grid
<br />Str. random
<br />Random grid
<br />Str. random
<br />©2008 International Society of Arboriculture
<br />
|