170 (1976 to present; the publication of the International Society of Arboriculture was re-named to the latter in 2006), Arboricultural Journal (1965 to present), Cit- ies and the Environment (2008 to present), and Urban Forestry & Urban Greening (2002 to present), since these are the journals most likely to have studies of interest. The date range for searching was “all time” for keyword searches, or the earliest publications for the comprehensive journal searches. This included material available online through December 2017. We carried out keyword and title searches using the following terms: tree mortality, tree survival, tree survivability, tree survivorship, tree death, tree removal, tree population projection, tree population model, tree establishment, and tree failure. These terms were joined by the search term “AND” with the words urban, city, street, and yard. After identifying initial articles, more potential articles were found using “backward chaining” (searching the literature cited in the starting manuscript, then moving back- ward through a chain of references) and “forward chaining” (finding articles which cite the starting manuscript, following the chain of references for- ward). Both “chaining” techniques can be successful for comprehensive literature searches (Booth 2008). Forward chaining was conducted using Google Scholar, as this search engine searches a broad range of literature and is more likely to locate “gray litera- ture” such as theses, extension articles, and confer- ence proceedings (Haddaway 2015). The inclusion of “gray literature” was evaluated by researchers on a case-by-case basis. Theses which were later published in journals were excluded. Studies were considered eligible for inclusion in our literature review if they: (1) examined tree mor- tality in landscaped or heavily built-up urban areas, such as trees in sidewalks, parking lots, yards, and manicured parks; and (2) were observational studies in real-world urban conditions. We excluded arbori- cultural planting experiments, such as tests of cultivar performance (e.g., Gerhold 2007), studies that focused completely on remnant or afforested forest fragments (e.g., Dislich and Pivello 2002), and stud- ies which stated mortality assumptions in projection models (e.g., McPherson 2008). We included some plot-based studies that examined tree mortality across an entire city, and therefore included wooded park lands, but we focused our discussion on the other land uses. ©2019 International Society of Arboriculture Hilbert et al: Urban Tree Mortality: A Literature Review Examining Mortality Information We targeted urban forestry studies that presented data on mortality or survival rates and the influential factors associated with mortality. These factors could have been examined either quantitatively or qualitatively. Field-based monitoring studies generally fell into two study design categories: repeated inventories of uneven-aged tree populations and planting cohort studies of even-aged trees. If a study examined a pop- ulation of trees of various size and age classes by comparing data from a current inventory with data from a prior inventory (either conducted by the same researchers or pulled from existing records), then we considered it to be a repeated inventory of uneven- aged trees. This category included monitoring i-Tree Eco style plots and repeated street tree inventories. If a study examined the survival and mortality of a group of trees planted around the same time, then we considered it to be a planting cohort study of rela- tively even-aged trees. In the context of planted urban trees, “age” means time since planting. Furthermore, some planting cohort studies examined groups of trees that were planted over a small range of years, which we considered to be a multi-year cohort. When we analyzed cohort studies, we considered the estab- lishment phase to be the first five years since planting and the post-establishment phase to be over five years since planting. Some urban tree mortality studies did not fall into either category and therefore could not be analyzed for mortality rates per se. However, since they still pertained to real-world mortality and pro- vided information on factors like human behavior and natural disasters, we included them in our review, the results, and discussion. These studies included sur- veys of residents and urban tree managers, one-time inventories following storm events, assessments of removal records, and a dendrochronology study. For all planting cohort study data we calculated, annual mortality, qannual lt , and cumulative survivorship, , were defined as where t is the number of years since planting and lt is the proportion of the original population remaining alive at time t (Roman and Scatena 2011; Roman et al. 2016). That original population is represented by baseline data, either the first inventory (for repeated inventory studies) or planting records (for planting cohort studies). For repeated inventory studies of
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
