Critical analysis of the potential of Psidium guajava cv Paluma (guava tree) for ozone biomonitoring under seasonal subtropical climate

Abstract

• Psidium guajava is an ozone subtropical bioindicator. • Air temperature and relative humidity are important variables for ozone bioindication. • Multivariate analysis is a useful tool for biomonitoring in marked seasonality environments. Psidium guajava cv. Paluma (guava) has been emphasized as good indicator of phytotoxic tropospheric O 3 levels in studies conducted under both controlled and field conditions. However, its performance as bioindicator was never critically evaluated under subtropical climate featured by well-defined wet and dry seasons. The current field study provided new contributions on this aspect. Guava cv. Paluma plants were exposed to ozone in several locations of a metropolitan region in SE Brazil - which is featured by well-defined climate seasonality - to collect data to be used to describe seasonal and spatial variations in leaf injury index and in other leaf traits. Multilinear biomonitoring model was also adjusted for different O 3 pollution descriptors (daily O 3 concentration, AOT40, SUM00 and SUM60), meteorological conditions (air temperature and relative humidity, solar radiation, and vapor pressure deficit) and morphological leaf traits (visible leaf injury, leaf area, leaf dry mass, and leaf mass per leaf area). Visible leaf injuries tended to be higher in locations near the industrial pole. Autumn was the season when guava plants recorded the highest leaf injury caused by O 3 . Linear regression analysis did not indicate significant association between different O 3 descriptors and mean leaf injury; however, multilinear regression analysis has shown that other independent variables increased the bioindicator model’s explicability (R 2 = 0.67). Mean leaf injury observed for the 3rd, 4th and 5th oldest guava cv. Paluma leaves were predicted based on combined effects of air temperature and relative humidity, as well as on O 3 expressed as AOT40. The other leaf traits were excluded from the multilinear model. The adjusted model was validated as significant tool for future studies about air quality in subtropical regions with seasonal climate.