280 (Angel 2011). Within the Morton landscape, a range of urban land-use categories were sampled: built (near buildings, park- ing lots), transportation (along major roadways – Interstate-88, IL-53), park (tree collections with grass or other landscaped un- derstory conditions), and forest (semi-natural forest conditions). To assess variation in species responses across these land-use categories, four tree species that are commonly planted in the urban forest were sampled: Kentucky coffeetree (Gymnocladus dioicus), eastern white pine (Pinus strobus), sugar maple (Acer saccharum), and tulip-tree (Liriodendron tulipifera). These spe- cies vary in their tolerance to drought conditions, with white pine and coffeetree both having considerable drought tolerance and sugar maple and tulip-tree both exhibiting susceptibility to low moisture conditions (Gilman and Watson 1993; Orwig and Abrams 1997; Caspersen and Kobe 2001). However, tolerance of low-moisture conditions is likely to be highly conditional on the characteristics of the individual plant (e.g., age, size, root–shoot ratio) and is likely to be site-dependent as well, making specific a priori predictions about drought impacts on species difficult. To examine tree growth response to historical climate, incre- ment cores were collected from at least 25 trees in each land-use category and from at least 25 trees of each species. However, not all cores were readable, and so sample sizes varied among both land-uses and species (Table 1). Cores were collected at 1.37 m and attempted to capture at least 20 years of growth increments. Increment cores were mounted on grooved wood blocks and sanded using progressively finer sandpaper to help distinguish rings. Annual growth increments were measured with a Velmex stage micrometer using Measure J2X software (Voortech 2005). Data Analysis Yearly growth for each tree was calculated by converting ring width to basal area increment (BAI) by back calculating basal area based on current diameter and annual ring widths. Total growth increment over the past 15 years (1997–2011) was calcu- lated for each tree. To account for the effect of tree size and age on growth in comparing across land-use categories, BAI for each tree was adjusted for starting diameter and tree age. A multiple regression equation explaining BAI as a function of age and start- ing diameter (in 1997) was fit to the data and an expected BAI value was calculated for each tree based on the resulting equation: [1] [-3.1619 • (Age)] + [-0.0045 • (Age • BA)] Actual total BAI for each tree over the 15-year study Expected BAI1997–2011 = 225.10 + 0.63 • (BA) + period was then scaled by this expected value to pro- duce a scaled BAI value using the following equation: [2] / (Expected BAI)2 Scaled BAI = Total BAI • [(Total BAI • Expected BAI) ] Built Forest Park Transportation Total Sugar maple 9 8 8 4 29 ©2013 International Society of Arboriculture Kentucky coffeetree 7 3 6 5 21 Fahey et al.: Tree Growth and Resilience to Extreme Drought Variation in growth was also assessed by calculating the coef- ficient of variation (CV; standard deviation/mean) of BAI for each tree over the same 15-year period. Both scaled BAI and CV of BAI were compared among land-use categories (Question 1) using anal- ysis of variance (ANOVA). All ANOVA analyses were conducted using PROC GLM in SAS v.9.2 (SAS-Institute-Inc. 2005) and Tukey-Kramer adjustments were applied to means comparisons. In order to assess differences among land-use categories in sensitivity of tree growth to environmental variation (Ques- tion 2), the correlation of BAI with annual and growing season (May–October) precipitation was determined for each tree. For the 15-year period of interest, annual and growing season pre- cipitation were determined based on climate records from the Wheaton climate monitoring station (Angel 2011). Correlation of BAI with precipitation was then calculated for each tree and compared among land-use categories and species using ANOVA. To evaluate variability in growth in relation to extreme drought (Question 3), researchers investigated tree growth responses to a drought that occurred in 2005 (considered one of the three most intense droughts in the historical record in Illi- nois; Kunkel et al. 2006). Statewide, an average of only 50.01 cm of precipitation fell during the period of March–October 2005 (compared with statewide normal of 74.22 cm; Kunkel et al. 2006). Drought resistance can be defined as the ability of a tree to maintain growth under drought conditions. A drought resistance index can be calculated by comparing growth in the drought year relative to average growth over the previous five-year period: BAI2005/mean BAI2000-04 (D’Amato et al. 2011). Resilience to drought is the ability of a tree to return growth to pre-drought levels in the period following the drought. A drought resilience index was defined as the average growth in the five years fol- lowing the drought relative to pre-drought growth: resilience index = mean BAI2006-10/ mean BAI2000-04 (D’Amato et al. 2011). For these indices, values <1 indicate a negative growth response in the year of the drought (resistance) or the five-year period following the drought (resilience) relative to pre-drought growth rates. Drought resistance and resilience indices were compared among land-use categories and species using ANOVA. RESULTS Growth differed strongly among land-use categories even after accounting for differences in age and diameter (F3, 112 = 6.49, p < = 12.97, p < 0.001), as tulip-tree and white pine, both shade-intolerant early- successional species, had greater growth rates in the park land-use than the other species (Figure 2a). Variation in growth was gener- ally relatively low (average CV = 0.49) and did not differ signifi- cantly among land-use categories (F3, 112 Table 1. Count of readable increment cores by land-use category and species. Land-use category Tulip-tree 4 8 10 7 29 White pine 11 8 9 9 37 Total 31 27 33 25 116 0.001; Figure 1a). The highest growth rates were seen in the park land-use category, with both forest and transportation having lower total growth (Figure 1a). There was also a significant inter- action between land-use and species in growth (F15, 100 = 1.94, p = 0.13; Figure
November 2013
| Title Name |
Pages |
Delete |
Url |
| Empty |
Search Text Block
Page #page_num
#doc_title
Hi $receivername|$receiveremail,
$sendername|$senderemail wrote these comments for you:
$message
$sendername|$senderemail would like for you to view the following digital edition.
Please click on the page below to be directed to the digital edition:
$thumbnail$pagenum
$link$pagenum
Your form submission was a success. You will be contacted by Washington Gas with follow-up information regarding your request.
This process might take longer please wait
