Testing the accuracy of resistance drilling to assess tree growth rate and the relationship to past climatic conditions

• Published in: Urban forestry & urban greening 2018-12, Vol.36, p.1-12
• Main Authors: Orozco-Aguilar, Luis, Nitschke, Craig R., Livesley, Stephen J., Brack, Cris, Johnstone, Denise
• Format: Article
• Language: English
• Subjects: Dendrochronology; Drilling profiles; Growth trends; Ring analysis; Urban forests
• ISSN: 1618-8667; 1610-8167
• DOI: 10.1016/j.ufug.2018.09.010

•Resistance drilling (RD) can assess mean ring width values to a comparable standard as conventional stem core analysis.•RD readings allowed the identification of pointer years at population level but could not detect event years at tree level.•The accuracy of RD for ring analysis weakened as tree ring width decreased below 1 mm and for mature/old trees.•RD can provide a chronology master series, but it may not be as robust as with a core sampling ring width series.•RD should not replace dendrochronology in climate-tree growth studies, unless RD accuracy can be improved and demonstrated. Assessing tree growth trends over time is a central but challenging aspect of urban forest management. The potential damage caused by invasive devices used in dendrochronological analysis is a common concern among urban foresters. Thus, the development of a less-invasive method for assessing tree growth rate faster that provides reliable results is clearly beneficial. In this study, resistance drilling (RD) profiles were compared with stem core assessments (Core) to estimate the growth rate of 78 trees of three species (Quercus robur, Ulmus procera, and Platanus x acerofolia). All studied trees were core-sampled in 2013 and then resistance drilled in 2015 at a stem height of 1–1.3 m in both north (N) and west axes (W). The dependency and accuracy of paired annual ring series (CORE measurements and Resi reading) were tested using ANOVA and regression analysis. In addition, point and event year tests were determined to confirm the accuracy of the RD to assess growth trends at both population and tree level. Growth series from both methods were cross-dated to test the reliability of RD to relate historical tree growth to past climatic conditions. ANOVA analysis confirmed that average ring width values and age of 70 out of 78 trees were statistically similar for both methods and similar for both sampled stem axes. Within each tree, regression analysis indicated significant correlation between cored ring datasets and paired resistance drilled ring datasets (R2 = 0.78–0.95, p