Differences in quantity and composition of leaf particulate matter and morphological structures in three evergreen trees and their association in Harbin, China

Urban trees have been assumed to effectively clean air particulate matter (PM), while the inter-species differences are not yet well defined, especially the PM chemical composition. In this study, PM from leaf surface and wax layer of 3 evergreen tree species (Juniper: Juniperus rigida; Black pine: Pinus tabuliformis var. mukdeais; Spruce: Picea koraiensis) were used for finding differences in PM adsorption and its compositional traits (characterized by X-ray diffraction, X-ray photoelectron spectroscopy, Fourier transform infrared spectrum and Inductively coupled plasma-optical emission spectrometry). Possible improvement in PM removal was also evaluated by a detail whole city tree census and different scenarios of species adjustment data. We found that: 1) the amount of PM on juniper leaves was 5.73 g m(-2), 2-2.5 fold higher than black pine and spruce (p < 0.05). Of them, 38.73%, 38.22%, and 23.11% were in the wax layer. 2) Compared with the explicit interspecies differences in PM quantity, more complex interspecies difference showed different patterns for different compositional traits. In general, leaf surface PM had higher O, Si, Al, Fe, N, Pb, Cu, Ni, Cr, and Cd, while the wax PM had higher C and Na contents (p < 0.05). 3) Association ordination found that the smaller leaf size, lower leaf water content, higher leaf area per unit mass, higher wax content, and larger stomatal openness aligned with the more PM adsorption by leaf, together with the higher amounts of C=O stretching, O, Si, Al, N, heavy metals of Pb, Cu, Ni, Cr, and Cd in PM. 4) Compared with the other 2 species, increase of juniper percentage in urban forests is more effective for maximizing PM removal from air, accompanying more heavy metal removal but less crystalized minerals in PM. Our findings highlight that proper species configuration in urban afforestation could maximize the air PM removal capacity. (C) 2019 Elsevier Ltd. All rights reserved.