Comparisons of PM2.5 mitigation with stand characteristics between evergreen Korean pine plantations and deciduous broad-leaved forests in the Republic of Korea.

Owing to industrialization and urbanization in recent decades, fine particulate matter (PM2.5) in the atmosphere has become a major environmental problem worldwide. This environmental issue pushed the use of forests as air filtering tools. However, there is a lack of continuous and long-term forest management to efficiently mitigate PM2.5. In this study, we assessed the potential of different forest types to control air pollution by measuring the seasonal PM2.5 concentrations inside and outside the forest for one year. In addition, the PM2.5 reduction efficiencies (PMREs) of two forest types were compared, and their relationship with stand characteristics was analyzed. The results showed that the average PMRE inside the forests was approximately 18.2%; the seasonal PMRE was highest in winter (approximately 28.1%) and lowest in summer (approximately 9.6%). The average PMRE of the Taehwa deciduous broad-leaved forest (TDF) (approximately 18.8%) was significantly higher than that of the Taehwa coniferous forest (TCF) (approximately 17.5%) (P < 0.001); differences were also observed seasonally. The PMRE in the TCF was higher in spring and summer (P < 0.001), while that in the TDF was higher in autumn and winter (P < 0.001). Furthermore, the PMRE in the TDF was negatively correlated with stand density (P = 0.003) and positively correlated with the average diameter at breast height (DBH) (P = 0.028). However, the PMRE in the TCF did not significantly correlate with stand characteristics. As such, the results of this study revealed the differences in PM2.5 mitigation according to stand characteristics, which should be considered in urban forest management.

Identifying and extracting bark key features of 42 tree species using convolutional neural networks and class activation mapping.

The significance of automatic plant identification has already been recognized by academia and industry. There were several attempts to utilize leaves and flowers for identification; however, bark also could be beneficial, especially for trees, due to its consistency throughout the seasons and its easy accessibility, even in high crown conditions. Previous studies regarding bark identification have mostly contributed quantitatively to increasing classification accuracy. However, ever since computer vision algorithms surpassed the identification ability of humans, an open question arises as to how machines successfully interpret and unravel the complicated patterns of barks. Here, we trained two convolutional neural networks (CNNs) with distinct architectures using a large-scale bark image dataset and applied class activation mapping (CAM) aggregation to investigate diagnostic keys for identifying each species. CNNs could identify the barks of 42 species with > 90% accuracy, and the overall accuracies showed a small difference between the two models. Diagnostic keys matched with salient shapes, which were also easily recognized by human eyes, and were typified as blisters, horizontal and vertical stripes, lenticels of various shapes, and vertical crevices and clefts. The two models exhibited disparate quality in the diagnostic features: the old and less complex model showed more general and well-matching patterns, while the better-performing model with much deeper layers indicated local patterns less relevant to barks. CNNs were also capable of predicting untrained species by 41.98% and 48.67% within the correct genus and family, respectively. Our methodologies and findings are potentially applicable to identify and visualize crucial traits of other plant organs.