PS 57-169
Quantifying urban forests: A pilot test for the Urban Tree Growth and Longevity working group's minimum data set and field guide

Wednesday, August 12, 2015
Exhibit Hall, Baltimore Convention Center
Lara A. Roman, Philadelphia Field Station, USDA Forest Service, Philadelphia, PA
Daniel R. Betz, Ecology and Evolution, Rutgers University, New Brunswick, NJ
Bryant Scharenbroch, University of Wisconsin-Stevens Point
Lee Mueller, Friends of Grand Rapids Parks
Johan Ostberg, Swedish University of Agricultural Sciences
Jess Sanders, Casey Trees
Andrew Koeser, University of Florida
Erik Desotelle, University of Wisconsin-Stevens Point
Background/Question/Methods

Can minimally trained individuals accurately collect basic urban forestry data? Much research has focused on the services provided by the urban forests, and many cities are currently looking for ways to better manage their trees. Urban tree management requires knowing what you have, and in many cities the number of trees present is infeasible to have professionals collecting all of the relevant data, therefore, cities are turning to volunteers and minimally trained individuals. The UTGL working group of the International Society of Arboriculture developed a field guide and minimum data set to standardize data and make it more useful to the long term studies of urban forests.  A pilot test of their field guide and collection protocols was run in six cities, to determine the viability of using minimally trained individuals. Twelve volunteers were recruited in each city, and the volunteers were trained on how to collect data and broken into 6 pairs, 3 of novice background and 3 of intermediate background.  They were then sent to measure 150 trees in their respective cities, and the data they collected was compared to data collected by experts who measured the same trees.  

Results/Conclusions

From the six cities involved data from two, Philadelphia, PA, and Lombard, IL has been compiled and shows pairs with a background classed as intermediate were better able to identify trees to the genus and species level, site type, transparency, and number of stems, while novice pairs did a better job determining land use, mortality, and wood condition. Dieback and diameter were recorded approximately as well by both intermediate and novice pairs.  The appropriate number of stems to measure on multistemed trees was poorly determined by both intermediate and novice groups.  The additional prior training had by the intermediate pairs did not lead to them dramatically outperforming the novice pairs. Both novice and intermediate pairs struggled with several of the factors being measured indicating the training protocols need to be adapted to address some of the more difficult variables, and less time could be spent on some of the more self-explanatory ones.  The condition variables of dieback, transparency, and wood condition need to be rethought.  Since diameter is so important new rules need to be constructed for dealing with multistemed trees and better training needs to be devised for accurate measurement.