Tropospheric Ozone Assessment Report: Tropospheric ozone from 1877 to 2016, observed levels, trends and uncertainties

Abstract

From the earliest observations of ozone in the lower atmosphere in the 19th century, both measurement methods and the portion of the globe observed have evolved and changed. These methods have different uncertainties and biases, and the data records differ with respect to coverage (space and time), information content, and representativeness. In this study, various ozone measurement methods and ozone datasets are reviewed and selected for inclusion in the historical record of background ozone levels, based on relationship of the measurement technique to the modern UV absorption standard, absence of interfering pollutants, representativeness of the well-mixed boundary layer and expert judgement of their credibility. There are significant uncertainties with the 19th and early 20th-century measurements related to interference of other gases. Spectroscopic methods applied before 1960 have likely underestimated ozone by as much as 11% at the surface and by about 24% in the free troposphere, due to the use of differing ozone absorption coefficients.There is no unambiguous evidence in the measurement record back to 1896 that typical mid-latitude background surface ozone values were below about 20 nmol mol -1 , but there is robust evidence for increases in the temperate and polar regions of the northern hemisphere of 30-70%, with large uncertainty, between the period of historic observations, 1896-1975, and the modern period (1990-2014). Independent historical observations from balloons and aircraft indicate similar changes in the free troposphere. Changes in the southern hemisphere are much less. Regional representativeness of the available observations remains a potential source of large errors, which are difficult to quantify.The great majority of validation and intercomparison studies of free tropospheric ozone measurement methods use ECC ozonesondes as reference. Compared to UV-absorption measurements they show a modest (~1-5% ±5%) high bias in the troposphere, but no evidence of a change with time. Umkehr, lidar, and FTIR methods all show modest low biases relative to ECCs, and so, using ECC sondes as a transfer standard, all appear to agree to within one standard deviation with the modern UV-absorption standard. Other sonde types show an increase of 5-20% in sensitivity to tropospheric ozone from 1970-1995. Biases and standard deviations of satellite retrieval comparisons are often 2-3 times larger than those of other free tropospheric measurements. The lack of information on temporal changes of bias for satellite measurements of tropospheric ozone is an area of concern for long-term trend studies.

Tropospheric ozone in CMIP6 Simulations

“…The models underestimate the magnitude of the observed reduction, and consequently, simulate a small growth here. pared to more accurate UV-absorption measurements (Tarasick et al, 2019). Measurement precision is ±3-5 % and the overall uncertainty in ozone concentration is less than 10 % in the troposphere (Kerr et al, 1994;Smit et al, 2007;Tarasick et al, 2016Tarasick et al, , 2019.…”

Global-scale distribution of ozone in the remote troposphere from ATom and HIPPO airborne field missions

“…All sites use electrochemical concentration cell (ECC) ozonesondes that rely on the potassium iodide electrochemical detection of O3, and which provide a vertical resolution of about 100 m (Komhyr, 1969). The associated uncertainty is usually ± (5-10 %) (Tarasick et al, 2019a;Thompson et al, 2019;Witte et al, 2018). ATom.…”

Statistical regularization for trend detection: An integrated approach for detecting long-term trends from sparse tropospheric ozone profiles

“…The IAGOS program has measured ozone worldwide since 1994, using instruments onboard commercial aircraft of internationally operating airlines (Marenco et al, 1998;Petzold et al, 2015, http://www.iagos.org). This analysis utilizes the same subset of IAGOS ozone profiles above Western Europe reported by Cooper et al (2019). Ozone is measured using a dual-beam UV-absorption monitor (time resolution of 4 seconds) with an accuracy estimated at about ± (2 nmol mol −1 + 2 %) (Thouret et al, 1998;Nédélec et al, 2015 (Johnson et al, 2002;Smit et al, 2007).…”

“…The ozonesondes have been launched on a weekly schedule since 1992 (Oltmans et al, 1996) with a lower frequency of approximately 2-4 profiles per month from 1982 to 1991. Compared to UV-absorption measurements, ECC ozonesondes show a modest (∼1-5% ± 5%) high bias in the troposphere, but no change over time (Tarasick et al, 2019 Using the interannual component in Figure 2(b) we calculate the trend value for each individual pressure layer. This integrated fit is then compared to the separated linear fit in each layer based on Eq (1) and the original monthly mean time series (with seasonal cycle and autocorrelation accounted for).…”

“…The vertical distributions of the Hilo (1982) and Trinidad Head (1998 ozone trends from the separated and integrated fits are shown in Figure 13, reported at a 20 hPa vertical resolution, as opposed to 50 hPa resolution for the IAGOS profiles above Europe and China. For reference, the trend estimation from the 1982-2018 surface ozone measurements at Mauna Loa Observatory (∼3400 m elevation; 60 km southwest of Hilo) was added to the Hilo plot (Cooper et al, 2019). The separated fit results in trends that can vary considerably from one layer to the next.…”