RT Journal Article
SR Electronic
T1 Selecting Reference Cities for i-Tree Streets
JF Arboriculture & Urban Forestry (AUF)
FD International Society of Arboriculture
SP 230
OP 240
DO 10.48044/jauf.2010.031
VO 36
IS 5
A1 E. Gregory McPherson
YR 2010
UL http://auf.isa-arbor.com/content/36/5/230.abstract
AB The i-Tree Streets (formerly STRATUM) computer program quantifies municipal forest structure, function, and value using tree growth and geographic data from sixteen U.S. reference cities, one for each of sixteen climate zones. Selecting the reference city that best matches a subject city is problematic when the subject city is outside the U.S., lays on the border between two climate zones, has a different climate, or tree species composition because of differences in elevation, urban morphology, and environmental quality. A systematic process for selecting the best match is described and illustrated for Lisbon, Portugal. Selection criteria are tree species composition, heating and cooling degree days, and annual precipitation. Raw and difference values for each criterion are normalized to range from 0 to 10 using linear interpolation. The coefficient for each criterion is weighted to reflect its relative importance. The Root Mean Square Error (RMSE) is calculated and the reference city with the lowest value is the best match for the subject city. The state of California’s reference cities of Modesto (RMSE = 2.41) and Claremont (2.71) proved to be the best match for Lisbon when coefficients were unequally weighted.