While often considered “open grown,” urban trees are often found in relatively close proximity to neighboring trees, buildings, and other elements of urban infrastructure. These spatial arrangements may provide wind protection during severe weather events such as hurricanes. Beyond this very local scale, urban tree abundance and condition are often influenced by the greater sociodemographic context of the neighborhood or community where they are found. In an effort to assess the impact of these external factors on wind firmness after Hurricane Irma impacted Florida in 2017, we revisited three urban areas that had previously been inventoried using sample plots prior to the storm. At each plot we assessed storm damage and characterized surrounding protective elements (i.e., buildings and other trees) using a combination of ground-based and aerial approaches. This was then paired with block group level sociodemographic data derived from the United States Census. Logistic regression results confirmed previous research, showing that partial and whole tree failure were significantly more likely with a larger stem diameter (p = 0.001) and perceived/previously documented lower wind firmness (p = 0.004). However, our results disagreed with previous studies linking species diversity and resilience, showing tree failure was significantly more likely with higher Shannon Diversity Index (p = 0.008). A comparative geographically weighted logistic regression model also found that the higher proportions of African American residents and Hispanic residents, and the median age of residents were significant predictors of less likely tree failure in 30 %, 36 %, and 20 % of plots, respectively. However, this evidence was weak, compared to that of tree- and plot-level effects. Despite being a key predictor of interest, we did not see any significant protective effect from neighboring trees or structures.

尽管通常被认为是“开放式种植”，但城市树木通常相对靠近邻近的树木，建筑物和城市基础设施的其他元素。这些空间布置可以在诸如飓风的恶劣天气事件期间提供防风。除了这个非常局部的规模之外，城市树木的丰度和状况通常还受到其所在社区或社区更大的社会人口统计学背景的影响。为了评估2017年飓风``艾尔玛''（Irma）袭击佛罗里达州之后这些外部因素对风稳固性的影响，我们重新评估了三个之前曾在暴风雨前使用样地进行盘点的城市地区。在每个图块中，我们评估了暴风雨的破坏并确定了周围的保护元素（例如，建筑物和其他树木）结合使用地面和空中方法。然后，将其与源自美国人口普查的部门群体级社会人口统计学数据配对。Logistic回归结果证实了先前的研究，结果表明，较大的茎直径（p = 0.001）和感知/先前记录的较低的风坚度（p = 0.004），部分和整个树的失败可能性更大。但是，我们的结果与先前有关物种多样性和复原力的研究不同意，研究表明，较高的香农多样性指数（p = 0.008）可能导致树木倒闭。比较地理加权逻辑回归模型也发现，非裔美国人和西班牙裔居民比例较高，居民年龄中位数分别是30％，36％和20％样地发生树木可能性较小的重要预测指标。但是，与树级效果和地块级效果相比，该证据还很薄弱。尽管是感兴趣的主要预测指标，但我们没有看到来自邻近树木或建筑物的任何显着保护作用。