A simple linear model for estimating ozone AOT40 at forest sites from raw passive sampling data

Area of Science:

• Environmental Science
• Atmospheric Chemistry
• Ecology

Background:

• Accurate assessment of ozone exposure is crucial for understanding its impact on forest ecosystems.
• Conventional ozone monitoring methods can be resource-intensive and logistically challenging, especially in remote forest areas.
• Passive samplers offer a cost-effective alternative for air quality monitoring but require robust methods for data interpretation.

Purpose of the Study:

• To develop and validate a rapid, empirical method for estimating weekly accumulated ozone exposure above a threshold of 40 ppb (AOT40) using passive ozone samplers.
• To assess the reliability of this new method by comparing its results with conventional monitoring data and established modeling techniques.
• To provide a practical tool for estimating forest ozone exposure where traditional methods are not feasible.

Main Methods:

• Development of a linear regression model based on three years of ozone concentration data from passive samplers in Trentino, northern Italy.
• Validation of the method using an independent dataset from passive sampler sites across Italy.
• Comparison of estimates with data from conventional ozone monitors and a modeling method based on hourly concentrations.

Main Results:

• The empirical method provided good weekly AOT40 estimates, with R(2) values ranging from 0.85 to 0.970 and RMSE values between 97 and 302 when compared to conventional monitors.
• Estimates from passive sampling were comparable to those from a modeling method (R(2) = 0.94).
• Independent testing yielded similar high correlations (0.86 ≤R(2)≤ 0.99), though errors accumulated when summing weekly estimates for the May-July period (median deviation of 11%).

Conclusions:

• The proposed linear regression method offers a rapid and practical approach for estimating weekly AOT40 in forest environments using passive ozone samplers.
• This method is suitable for situations where conventional monitoring or complex modeling is not feasible.
• While useful for weekly estimates, users should be aware of potential error accumulation when calculating cumulative seasonal AOT40.