2This study compiled recent inventory data from 929,823 street trees in 50 cities to determine 3 trends in tree number and density, identify priority investments and create baseline data against 4 which the efficacy of future practices can be evaluated. The number of street trees increased 5 from 5.9 million in 1988 to 9.1 million in 2014, about one for every four residents. Street tree 6 density declined from 65.6 to 46.6 trees per km, nearly a 30% drop. City streets are at 36.3% of 7 full stocking. State-wide, only London planetree (Platanus x hispanica) comprises over 10% of 8 the total, suggesting good state-wide species diversity. However, at the city scale, 39 9 communities were overly reliant on a single species. The state's street trees remove 567,748 t 10

Rainfall interception research in forest ecosystems usually focuses on interception by either tree crown or leaf litter, although the 2 components interact when rainfall occurs. A process‐based study was conducted to jointly measure rainfall interception by crown and litter and the interaction between the 2 interception processes for 4 tree species (Platycladus orientalis and Pinus tabulaeformis represented needle‐leaf species, and Quercus variabilis and Acer truncatum represented broadleaf species) at 3 simulated rainfall intensities (10, 50, and 100 mm hr−1). Results indicated that (a) crown and litter interception processes incorporated 3 phases: the dampening phase, the steady saturation phase, and the postrainfall drainage phase, but the dampening phase for litter interception usually lasted 30 min longer than for crown interception; (b) the maximum and minimum interception storage (Cmax and Cmin) for the crown were 0.63 and 0.36 mm on average, and litter Cmax and Cmin were 5.38 and 2.36 mm, respectively; (c) generally, crown and litter Cmax and Cmin increased when gross precipitation increased significantly (p < .05) from 10 to 100 mm; and (4) crown Cmax and Cmin for needle‐species were 1.8 and 1.2 times larger than broadleaf species, whereas broadleaf litter showed the opposite, its Cmax and Cmin were 2.0 and 1.6 times larger than needle‐leaf litter on average; however, no significant differences were observed in crown and litter Cmax and Cmin between species on per leaf area and litter thickness basis. Results were normalized by total leaf area and litter thickness to provide a way to scale up from young trees to mature forests. Overall, rainfall interception was affected by biotic and abiotic factors together and could be quantified via multiple linear regression functions.

Tree survival is a performance metric for urban forestry initiatives, and an understanding of the factors that influence mortality can help managers target resources and enhance survival. Furthermore, urban tree planting investments depend on tree survival to maximize ecosystem services. In this literature review, we categorized factors commonly associated with urban tree mortality and summarized mortality rates published in 56 studies, focusing on studies of trees along streets, in yards, and in landscaped parks. Study designs included quantitative field monitoring of uneven-aged tree populations and tracking planting cohorts of even-aged trees, as well as qualitative analyses. Annual mortality rates ranged from 0.6 to 68.5% for cohort studies and 0 to 30% for repeated inventories of uneven-aged trees. The 1st, 2nd, and 3rd quartiles of annual mortality were 2.8 to 3.8%, 4.4 to 6.5%, and 7.1 to 9.3% for planting cohorts, and 1.6%, 2.3 to 2.6%, and 3.0 to 3.3% for repeated inventories of uneven-aged trees (ranges reflect studies that reported a range for the time period or mortality rate). For cohort studies, annual mortality tended to be highest during the first five years after planting. The most commonly cited biophysical factors associated with mortality were taxa (15 articles), tree size/age (13 articles), and site characteristics (12 articles). The most commonly cited human-related factors were stewardship, maintenance, and vandalism (15 articles). More long-term studies are needed to investigate how site characteristics influence mortality, including rarely examined soil and microclimate characteristics. Future research should also examine institutional structures related to mortality outcomes, as well as parcel-level sociodemographic factors and resident behaviors.

We introduce an approach for updating older tree inventories with geographic coordinates using street-level panorama images and a global optimization framework for tree instance matching. Geolocations of trees in inventories until the early 2000s where recorded using street addresses whereas newer inventories use GPS. Our method retrofits older inventories with geographic coordinates to allow connecting them with newer inventories to facilitate long-term studies on tree mortality etc. What makes this problem challenging is the different number of trees per street address, the heterogeneous appearance of different tree instances in the images, ambiguous tree positions if viewed from multiple images and occlusions. To solve this assignment problem, we (i) detect trees in Google street-view panoramas using deep learning, (ii) combine multi-view detections per tree into a single representation, (iii) and match detected trees with given trees per street address with a global optimization approach. Experiments for > 50000 trees in 5 cities in California, USA, show that we are able to assign geographic coordinates to 38 % of the street trees, which is a good starting point for long-term studies on the ecosystem services value of street trees at large scale.

An experiment was conducted to concentrate on the rainfall interception process of individual trees for four common species in Beijing, China, which included needle species (Platycladus orientalis and Pinus tabulaeformis) and broadleaf species (Quercus variabilis and Acer truncatum). Two types of interception storages, the maximum (Cmax) and the minimum interception storage (Cmin), were examined at four simulated rainfall intensities (from 11.7 to 78.5 mm h−1). Results showed that an average of 91% of Cmax for all the species was intercepted during the first 10 minutes of rainfall, while 45% of Cmax drained off after rainfall cessation. Leaf area index (LAI) and leaf area (LA) were significantly correlated (p &lt; 0.05) with Cmax and Cmin, while such significant correlations were not found between rainfall intensity and Cmax and Cmin. Average Cmax and Cmin across all the species corresponded to 3 and 1% of gross rainfall. Mean Cmax and Cmin of the needle species were 3.0 and 1.8 times larger than that for the broadleaf ones. Results revealed that interception was a dynamic process which encompassed three phases. In addition, LAI and LA were valid predictors of interception in small trees, and deserve further test in forest stands.