180 (FIA) systems (Nowak et al. 2004; Staudhammer et al. 2011; Lawrence et al. 2012; Lima et al. 2013; Roman et al. 2014a), which could explain the preva- lence of variables like land use (and the specific land use categories observed). While these systems are useful for outlining potential variables to measure, researchers should think critically about which vari- ables are best for their specific mortality studies. For instance, locally-relevant land use and planting site categories could be more informative. Tree health and vigor evaluations, as well as soil characteristics, might be uncommon in urban tree mortality studies because they require more time, training, and equip- ment to measure. There is ample evidence that soil quality and available soil volume are critical to tree growth and health (e.g., Urban 2008; Rahman et al. 2013; Layman et al. 2016; Scharenbroch et al. 2017). Yet only three of the mortality studies we reviewed (Impens and Delcarte 1979; Lawrence et al. 2013; Koeser et al. 2014) examined soil characteristics in relationship to growth, and none of the studies tested whether soil properties were associated with mortal- ity and survival. Additionally, tree condition and health evaluations are not well-defined and consis- tently applied (Bond 2010). In terms of human fac- tors influencing mortality, parcel-level ownership data (Roman et al. 2014b; Ko et al. 2015a) and con- struction or renovation permitting data (Steenberg et al. 2017) show promise for understanding the process of tree mortality, yet these data sets have thus far been rarely applied to statistical modeling of urban tree mortality. Qualitative Analyses of Factors Associated with Mortality for Field-Based Monitoring Studies Some of the studies that reported mortality data from field-based monitoring studies drew conclusions about influential factors based on qualitative data and observations of trends (e.g., Rhoads et al. 1981; Sklar and Ames 1985; Polanin 1991; Richardson and Shackleton 2014; Roman et al. 2015), and many of these findings complement research with statistical analyses. Rhoads et al. (1981) studied a cohort of street trees in Philadelphia, PA and determined the cumulative survivorship to be approximately 85% for each species, therefore concluding that species is not a significant factor. This study contradicts the many other studies that found taxa to be significant, but the following were in more agreement with the studies that used statistical analysis. Ip (1996) documented ©2019 International Society of Arboriculture Hilbert et al: Urban Tree Mortality: A Literature Review the mortality of a cohort of community-planted trees in the Northwest Region of Canada, gathering infor- mation on the cause of mortality or injury to seed- lings when they were encountered. Approximately half of the urban and rural projects reported damage to seedlings by people, lawn care equipment, and snowmobiles, so Ip (1996) concluded that smaller tree size could be associated with higher mortality. Sullivan (2004) studied trees planted by a nonprofit in San Francisco, CA, and found that many residents and neighbors cited vandalism as a reason for tree death or removal. Richardson and Shackleton (2014) assessed the condition of newly planted street trees in eleven towns in Eastern Cape, South Africa, in order to understand more about vandalism, finding 42% of recently planted street trees totally snapped, and no difference between snapped trunks for trees with or without protective structures. In a case study of tree planting programs with high survival rates in East Palo Alto and Philadelphia, Roman et al. (2015) con- cluded that appropriate species selection and planting techniques, small geographic areas, and time-inten- sive maintenance explained low annual mortality during establishment, ranging from 0.6 to 4.6%. They also found that for the planting project with the low- est published urban tree mortality, East Palo Alto, the few instances of tree deaths were attributed to car accidents and site conditions. Other Study Designs The aforementioned planting cohort studies (Appendix Table 1) clearly examined changes in a cohort from time of planting to the time of monitoring. Similarly, the repeated inventory studies (Appendix Table 2) examined changes between inventories conducted at two different times. Those are all field-based monitoring that produced mortality data based on analyses of change over time. However, some articles used other study designs and more indirect measures of mortality. We reviewed fourteen studies that made use of data from a single point in time. These studies included surveys of homeowners or other individuals who could pro- vide indirect information about tree mortality, one-time inventories after storms or other major disturbances, and other unique designs. These studies did not have before-and-after data that can be used to estimate rates of loss, but they provide insight into associated factors and are therefore included in the review. Three of these fourteen studies conducted surveys of individual people in order to learn more about
September 2019
| 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
