Suggested key variables for assessment of soil quality in urban roadside tree systems

Urban roadside soils are important growth media for roadside trees. However, typical assessment variables are limited in describing the characteristics of roadside soils. We assessed the characteristics of roadside soils using the pre- and new suggested variables and recommended optimal soil variables that are representative of roadside tree health. Seventy-three roadside soils were collected for measurement, while six urban forest soils were prepared as a control. Samples were used to evaluate both pre-suggested and new variables. The former included bulk density, penetration resistance (PR), pH, organic matter (OM), fluorescein diacetate (FDA) activity, and respiration. To improve the pre-suggested variables, we modified the bulk density using PR and investigated the elemental ratios and stable isotopic signatures of particulate organic matter (POM). Two criteria were used to select the variables for urban roadside soils: (1) the variable should identify distinct characteristics of roadside and urban forest soils and (2) the variable should have a high correlation with urban tree health variables: leaf chlorophyll content and tree vigor. The bulk density measured using the conventional method underestimated soil compaction because obtaining intact cores was challenging. The modified bulk density (BDmodified) obtained from the soil PR is suggested to better represent soil compaction. The roadside soils were affected by de-icing materials, construction debris, and atmospheric alkali particles, which increased the soil pH. The unexpectedly higher OM contents in the roadside soils, where tree origins are limited, possibly due to soil OM sources such as vehicular emissions, animal excreta, and sewer flooding. These OM sources may alter the C/H ratio (POM-C/H) and the stable isotopic signature of POM, leading to OM quality changes. Soil respiration better reflected the changes in the microbial activity of the roadside soils, rather than FDA activity. The newly suggested soil variables, BDmofieid, pH, POM-C/H, and RES, were significantly correlated with leaf chlorophyll content and tree vigor (P < 0.05). Using a multiple regression analysis, the newly suggested set of soil variables, including the BDmodified, soil pH, POM-C/H, and soil respiration, showed high predictive power for the growth of urban roadside trees. Future studies should apply these variables to other cities or broader areas and confirm their predictive ability regarding the health of roadside trees.