scholarly journals ML-TOMCAT: machine-learning-based satellite-corrected global stratospheric ozone profile data set from a chemical transport model

2021 ◽  
Vol 13 (12) ◽  
pp. 5711-5729
Author(s):  
Sandip S. Dhomse ◽  
Carlo Arosio ◽  
Wuhu Feng ◽  
Alexei Rozanov ◽  
Mark Weber ◽  
...  
Keyword(s):  
Transport Model ◽  
Stratospheric Ozone ◽  
Chemical Transport ◽  
Data Sets ◽  
Chemical Transport Model ◽  
Profile Data ◽  
Data Set ◽  
Ozone Profile ◽  
Satellite Instruments

Abstract. High-quality stratospheric ozone profile data sets are a key requirement for accurate quantification and attribution of long-term ozone changes. Satellite instruments provide stratospheric ozone profile measurements over typical mission durations of 5–15 years. Various methodologies have then been applied to merge and homogenise the different satellite data in order to create long-term observation-based ozone profile data sets with minimal data gaps. However, individual satellite instruments use different measurement methods, sampling patterns and retrieval algorithms which complicate the merging of these different data sets. In contrast, atmospheric chemical models can produce chemically consistent long-term ozone simulations based on specified changes in external forcings, but they are subject to the deficiencies associated with incomplete understanding of complex atmospheric processes and uncertain photochemical parameters. Here, we use chemically self-consistent output from the TOMCAT 3-D chemical transport model (CTM) and a random-forest (RF) ensemble learning method to create a merged 42-year (1979–2020) stratospheric ozone profile data set (ML-TOMCAT V1.0). The underlying CTM simulation was forced by meteorological reanalyses, specified trends in long-lived source gases, solar flux and aerosol variations. The RF is trained using the Stratospheric Water and OzOne Satellite Homogenized (SWOOSH) data set over the time periods of the Microwave Limb Sounder (MLS) from the Upper Atmosphere Research Satellite (UARS) (1991–1998) and Aura (2005–2016) missions. We find that ML-TOMCAT shows excellent agreement with available independent satellite-based data sets which use pressure as a vertical coordinate (e.g. GOZCARDS, SWOOSH for non-MLS periods) but weaker agreement with the data sets which are altitude-based (e.g. SAGE-CCI-OMPS, SCIAMACHY-OMPS). We find that at almost all stratospheric levels ML-TOMCAT ozone concentrations are well within uncertainties of the observational data sets. The ML-TOMCAT (V1.0) data set is ideally suited for the evaluation of chemical model ozone profiles from the tropopause to 0.1 hPa and is freely available via https://doi.org/10.5281/zenodo.5651194 (Dhomse et al., 2021).

scholarly journals ML-TOMCAT: Machine-Learning-Based Satellite-Corrected Global Stratospheric Ozone Profile Dataset from a Chemical Transport Model

2021 ◽  
Author(s):  
Sandip S. Dhomse ◽  
Carlo Arosio ◽  
Wuhu Feng ◽  
Alexei Rozanov ◽  
Mark Weber ◽  
...  
Keyword(s):  
Transport Model ◽  
Stratospheric Ozone ◽  
Chemical Transport ◽  
Chemical Transport Model ◽  
Vertical Coordinate ◽  
Ozone Profile ◽  
Minimal Data ◽  
Almost All ◽  
Satellite Instruments

Abstract. High quality stratospheric ozone profile datasets are a key requirement for accurate quantification and attribution of long-term ozone changes. Satellite instruments obtain stratospheric ozone profile measurements over typical mission durations of 5–15 years. Various methodologies have then been applied to merge and homogenise the different satellite data in order to create longer term observation-based ozone profile datasets with minimal data gaps. However, individual satellite instruments use different measurement methods, sampling patterns and retrieval algorithms which complicate the merging of these different datasets. In contrast, atmospheric chemical models can produce chemically consistent long-term ozone simulations based on specified changes in external forcings, but they are subject to the deficiencies associated with incomplete understanding of complex atmospheric processes and uncertain photochemical parameters. Here, we use chemically self-consistent output from the TOMCAT 3-D chemical transport model (CTM) and a Random-Forest (RF) ensemble learning method to create a merged 42-year (1979–2020) stratospheric ozone profile dataset (ML-TOMCAT V1.0). The underlying CTM simulation was forced by meteorological reanalyses, specified trends in long-lived source gas, solar flux and aerosol variations. The RF is trained using the Stratospheric Water and OzOne Satellite Homogenized (SWOOSH) dataset over the time periods of the Microwave Limb Sounder (MLS) from the Upper Atmosphere Research Satellite (UARS) (1991–1998) and Aura (2005–2016) missions. We find that ML-TOMCAT shows excellent agreement with available independent satellite-based datasets which use pressure as the vertical coordinate (e.g. GOZCARDS, SWOOSH for non-MLS periods) but weaker agreement with the datasets which are height-based (e.g. SAGE–CCI–OMPS, SCIAMACHY-OMPS). We find that at almost all stratospheric levels ML-TOMCAT ozone concentrations are well within uncertainties in the observational datasets. The ML-TOMCAT dataset is thus ideally suited for the evaluation of model ozone profiles from the tropopause to 0.1 hPa. ML-TOMCAT data is freely available via https://zenodo.org/record/4997959#.YNzleUlKiUk (Dhomse et al., 2021).

Machine-Learning-Based Satellite-Corrected Global Stratospheric Ozone Profile Dataset from a Chemical Transport Model

2021 ◽  
Author(s):  
Sandip Dhomse ◽  
Martyn Chipperfield
Keyword(s):  
Machine Learning ◽  
Transport Model ◽  
Stratospheric Ozone ◽  
Chemical Transport ◽  
Recovery Phase ◽  
Horizontal Resolution ◽  
Chemical Transport Model ◽  
High Quality ◽  
Ozone Profile

<p>High quality global ozone profile datasets are necessary to monitor changes in stratospheric ozone. Hence, various methodologies have been used to merge and homogenise different satellite datasets in order to create long-term observation-based ozone profile datasets with minimal data gaps. However, individual satellite instruments use different measurement methods and retrieval algorithms that complicate the merging of these different datasets. Furthermore, although atmospheric chemical models are able to simulate chemically consistent long-term datasets, they are prone to the deficiencies associated with the computationally expensive processes that are generally represented by simplified parameterisations or uncertain parameters.</p><p>Here, we use chemically consistent output from a 3-D Chemical Transport Model (CTM, TOMCAT) and an ensemble of three machine learning (ML) algorithms (Adaboost, GradBoost, Random Forest), to create a 42-year (1979-2020) stratospheric ozone profile dataset. The ML algorithms are primarily trained using the Stratospheric Water and OzOne Satellite Homogenized (SWOOSH) dataset by selecting the UARS-MLS (1992-1998) and AURA-MLS (2005-2019) time periods. This ML-corrected version of monthly mean zonal mean TOMCAT (hereafter ML-TOMCAT) ozone profile data is available at both pressure (1000 hPa - 1 hPa) and geometric height (surface to 50 km) levels at about 2.5 degree horizontal resolution.</p><p>We will present a detailed evaluation of ML-TOMCAT profiles against range of merged satellite datasets (e.g. GOZCARDS, SAGE-CCI-OMPS, and BVertOzone) as well high quality solar occultation observations (e.g. SAGE-II v7.0 (1984-2005), HALOE v19 (1991-2005) and ACE v4.1 (2004-2020). Overall, ML-TOMCAT shows good agreement with the evaluation datasets but with poorer agreement at low latitudes. We also show that, as in different merged satellite data sets, ML-algorithms show larger spread in the tropical middle stratosphere. Finally, we will present a trend analysis from TOMCAT and ML-TOMCAT profiles for the post-1998 ozone recovery phase.</p>

scholarly journals Solar response in tropical stratospheric ozone: a 3-D chemical transport model study using ERA reanalyses

2011 ◽  
Vol 11 (24) ◽  
pp. 12773-12786 ◽  
Author(s):  
S. Dhomse ◽  
M. P. Chipperfield ◽  
W. Feng ◽  
J. D. Haigh
Keyword(s):  
Standard Model ◽  
Transport Model ◽  
Stratospheric Ozone ◽  
Chemical Transport ◽  
Stratospheric Aerosol ◽  
Data Sets ◽  
Chemical Transport Model ◽  
Model Simulations ◽  
Model Study ◽  
Significant Difference

Abstract. We have used an off-line 3-D chemical transport model (CTM) to investigate the 11-yr solar cycle response in tropical stratospheric ozone. The model is forced with European Centre for Medium-Range Weather Forecasts (ECMWF) (re)analysis (ERA-40/operational and ERA-Interim) data for the 1979–2005 time period. We have compared the modelled solar response in ozone to observation-based data sets that are constructed using satellite instruments such as Total Ozone Mapping Spectrometer (TOMS), Solar Backscatter UltraViolet instrument (SBUV), Stratospheric Aerosol and Gas Experiment (SAGE) and Halogen Occultation Experiment (HALOE). A significant difference is seen between simulated and observed ozone during the 1980s, which is probably due to inhomogeneities in the ERA-40 reanalyses. In general, the model with ERA-Interim dynamics shows better agreement with the observations from 1990 onwards than with ERA-40. Overall both standard model simulations are partially able to simulate a "double peak"-structured ozone solar response with a minimum around 30 km, and these are in better agreement with HALOE than SAGE-corrected SBUV (SBUV/SAGE) or SAGE-based data sets. In the tropical lower stratosphere (TLS), the modelled solar response with time-varying aerosols is amplified through aliasing with a volcanic signal, as the model overestimates ozone loss during high aerosol loading years. However, the modelled solar response with fixed dynamics and constant aerosols shows a positive signal which is in better agreement with SBUV/SAGE and SAGE-based data sets in the TLS. Our model simulations suggests that photochemistry contributes to the ozone solar response in this region. The largest model-observation differences occur in the upper stratosphere where SBUV/SAGE and SAGE-based data show a significant (up to 4%) solar response whereas the standard model and HALOE do not. This is partly due to a positive solar response in the ECMWF upper stratospheric temperatures which reduces the modelled ozone signal. The large positive upper stratospheric solar response seen in SBUV/SAGE and SAGE-based data can be reproduced in model runs with fixed dynamical fields (i.e. no inter-annual meteorological changes). As these runs effectively assume no long-term temperature changes (solar-induced or otherwise), it should provide an upper limit of the ozone solar response. Overall, full quantification of the solar response in stratospheric ozone is limited by differences in the observed data sets and by uncertainties in the solar response in stratospheric temperatures.

scholarly journals Past changes in the vertical distribution of ozone – Part 1: Measurement techniques, uncertainties and availability

2014 ◽  
Vol 7 (5) ◽  
pp. 1395-1427 ◽  
Author(s):  
B. Hassler ◽  
I. Petropavlovskikh ◽  
J. Staehelin ◽  
T. August ◽  
P. K. Bhartia ◽  
...  
Keyword(s):  
Stratospheric Ozone ◽  
Measurement Techniques ◽  
United Nations Environment Programme ◽  
Critical Examination ◽  
Profile Measurement ◽  
Data Sets ◽  
Data Set ◽  
Ozone Profile ◽  
Measurement Uncertainties

Abstract. Peak stratospheric chlorofluorocarbon (CFC) and other ozone depleting substance (ODS) concentrations were reached in the mid- to late 1990s. Detection and attribution of the expected recovery of the stratospheric ozone layer in an atmosphere with reduced ODSs as well as efforts to understand the evolution of stratospheric ozone in the presence of increasing greenhouse gases are key current research topics. These require a critical examination of the ozone changes with an accurate knowledge of the spatial (geographical and vertical) and temporal ozone response. For such an examination, it is vital that the quality of the measurements used be as high as possible and measurement uncertainties well quantified. In preparation for the 2014 United Nations Environment Programme (UNEP)/World Meteorological Organization (WMO) Scientific Assessment of Ozone Depletion, the SPARC/IO3C/IGACO-O3/NDACC (SI2N) Initiative was designed to study and document changes in the global ozone profile distribution. This requires assessing long-term ozone profile data sets in regards to measurement stability and uncertainty characteristics. The ultimate goal is to establish suitability for estimating long-term ozone trends to contribute to ozone recovery studies. Some of the data sets have been improved as part of this initiative with updated versions now available. This summary presents an overview of stratospheric ozone profile measurement data sets (ground and satellite based) available for ozone recovery studies. Here we document measurement techniques, spatial and temporal coverage, vertical resolution, native units and measurement uncertainties. In addition, the latest data versions are briefly described (including data version updates as well as detailing multiple retrievals when available for a given satellite instrument). Archive location information for each data set is also given.

scholarly journals SI<sup>2</sup>N overview paper: ozone profile measurements: techniques, uncertainties and availability

2013 ◽  
Vol 6 (6) ◽  
pp. 9857-9938 ◽  
Author(s):  
B. Hassler ◽  
I. Petropavlovskikh ◽  
J. Staehelin ◽  
T. August ◽  
P. K. Bhartia ◽  
...  
Keyword(s):  
Stratospheric Ozone ◽  
Measurement Techniques ◽  
United Nations Environment Programme ◽  
Critical Examination ◽  
Profile Measurement ◽  
Data Sets ◽  
Data Set ◽  
Ozone Profile ◽  
Measurement Uncertainties

Abstract. Peak stratospheric chlorofluorocarbon (CFC) and other ozone depleting substance (ODS) concentrations were reached in the mid to late 1990s. Detection and attribution of the expected recovery of the stratospheric ozone layer in an atmosphere with reduced ODSs as well as efforts to understand the evolution of stratospheric ozone in the presence of increasing greenhouse gases are key current research topics. These require a critical examination of the ozone changes with an accurate knowledge of the spatial (geographical and vertical) and temporal ozone response. For such an examination, it is vital that the quality of the measurements used be as high as possible and measurement uncertainties well quantified. In preparation for the 2014 United Nations Environment Programme (UNEP)/World Meteorological Organization (WMO) Scientific Assessment of Ozone Depletion, the SPARC/IO3C/IGACO-O3/NDACC (SI2N) initiative was designed to study and document changes in the global ozone profile distribution. This requires assessing long-term ozone profile data sets in regards to measurement stability and uncertainty characteristics. The ultimate goal is to establish suitability for estimating long-term ozone trends to contribute to ozone recovery studies. Some of the data sets have been improved as part of this initiative with updated versions now available. This summary presents an overview of stratospheric ozone profile measurement data sets (ground- and satellite-based) available for ozone recovery studies. Here we document measurement techniques, spatial and temporal coverage, vertical resolution, native units and measurement uncertainties. In addition, the latest data versions are briefly described (including data version updates as well as detailing multiple retrievals when available for a given satellite instrument). Archive location information is for each data set is also given.

scholarly journals Trends in Stratospheric Ozone: Lessons Learned from a 3D Chemical Transport Model

2006 ◽  
Vol 63 (3) ◽  
pp. 1028-1041 ◽  
Author(s):  
Richard S. Stolarski ◽  
Anne R. Douglass ◽  
Stephen Steenrod ◽  
Steven Pawson
Keyword(s):  
Time Series ◽  
Solar Cycle ◽  
Ultraviolet Radiation ◽  
Boundary Conditions ◽  
Interannual Variability ◽  
Transport Model ◽  
Stratospheric Ozone ◽  
Chemical Transport ◽  
Chemical Transport Model ◽  
The Past

Abstract Stratospheric ozone is affected by external factors such as chlorofluorcarbons (CFCs), volcanoes, and the 11-yr solar cycle variation of ultraviolet radiation. Dynamical variability due to the quasi-biennial oscillation and other factors also contribute to stratospheric ozone variability. A research focus during the past two decades has been to quantify the downward trend in ozone due to the increase in industrially produced CFCs. During the coming decades research will focus on detection and attribution of the expected recovery of ozone as the CFCs are slowly removed from the atmosphere. A chemical transport model (CTM) has been used to simulate stratospheric composition for the past 30 yr and the next 20 yr using 50 yr of winds and temperatures from a general circulation model (GCM). The simulation includes the solar cycle in ultraviolet radiation, a representation of aerosol surface areas based on observations including volcanic perturbations from El Chichon in 1982 and Pinatubo in 1991, and time-dependent mixing ratio boundary conditions for CFCs, halons, and other source gases such as N2O and CH4. A second CTM simulation was carried out for identical solar flux and boundary conditions but with constant “background” aerosol conditions. The GCM integration included an online ozonelike tracer with specified production and loss that was used to evaluate the effects of interannual variability in dynamics. Statistical time series analysis was applied to both observed and simulated ozone to examine the capability of the analyses for the determination of trends in ozone due to CFCs and to separate these trends from the solar cycle and volcanic effects in the atmosphere. The results point out several difficulties associated with the interpretation of time series analyses of atmospheric ozone data. In particular, it is shown that lengthening the dataset reduces the uncertainty in derived trend due to interannual dynamic variability. It is further shown that interannual variability can make it difficult to accurately assess the impact of a volcanic eruption, such as Pinatubo, on ozone. Such uncertainties make it difficult to obtain an early proof of ozone recovery in response to decreasing chlorine.

scholarly journals Tagged ozone mechanism for MOZART-4, CAM-chem and other chemical transport models

2012 ◽  
Vol 5 (6) ◽  
pp. 1531-1542 ◽  
Author(s):  
L. K. Emmons ◽  
P. G. Hess ◽  
J.-F. Lamarque ◽  
G. G. Pfister
Keyword(s):  
Transport Model ◽  
Stratospheric Ozone ◽  
Chemical Transport ◽  
Chemical Mechanism ◽  
Chemical Transport Model ◽  
Transport Models ◽  
Chemical Tracers ◽  
Source Contributions ◽  
The Difference ◽  
No Emissions

Abstract. A procedure for tagging ozone produced from NO sources through updates to an existing chemical mechanism is described, and results from its implementation in the Model for Ozone and Related chemical Tracers (MOZART-4), a global chemical transport model, are presented. Artificial tracers are added to the mechanism, thus, not affecting the standard chemistry. The results are linear in the troposphere, i.e., the sum of ozone from individual tagged sources equals the ozone from all sources to within 3% in zonal mean monthly averages. In addition, the tagged ozone is shown to equal the standard ozone, when all tropospheric sources are tagged and stratospheric input is turned off. The stratospheric ozone contribution to the troposphere determined from the difference between total ozone and ozone from all tagged sources is significantly less than estimates using a traditional stratospheric ozone tracer (8 vs. 20 ppbv at the surface). The commonly used technique of perturbing NO emissions by 20% in a region to determine its ozone contribution is compared to the tagging technique, showing that the tagged ozone is 2–4 times the ozone contribution that was deduced from perturbing emissions. The ozone tagging described here is useful for identifying source contributions based on NO emissions in a given state of the atmosphere, such as for quantifying the ozone budget.

scholarly journals Stratospheric O3 changes during 2001–2010: the small role of solar flux variations in a chemical transport model

2013 ◽  
Vol 13 (19) ◽  
pp. 10113-10123 ◽  
Author(s):  
S. S. Dhomse ◽  
M. P. Chipperfield ◽  
W. Feng ◽  
W. T. Ball ◽  
Y. C. Unruh ◽  
...  
Keyword(s):  
Transport Model ◽  
Chemical Transport ◽  
Solar Flux ◽  
Data Sets ◽  
Chemical Transport Model ◽  
Weather Forecasts ◽  
Flux Data ◽  
Broadband Emission ◽  
Short Time Span ◽  
Modelling Studies

Abstract. Solar spectral fluxes (or irradiance) measured by the SOlar Radiation and Climate Experiment (SORCE) show different variability at ultraviolet (UV) wavelengths compared to other irradiance measurements and models (e.g. NRL-SSI, SATIRE-S). Some modelling studies have suggested that stratospheric/lower mesospheric O3 changes during solar cycle 23 (1996–2008) can only be reproduced if SORCE solar fluxes are used. We have used a 3-D chemical transport model (CTM), forced by meteorology from the European Centre for Medium-Range Weather Forecasts (ECMWF), to simulate middle atmospheric O3 using three different solar flux data sets (SORCE, NRL-SSI and SATIRE-S). Simulated O3 changes are compared with Microwave Limb Sounder (MLS) and Sounding of the Atmosphere using Broadband Emission Radiometry (SABER) satellite data. Modelled O3 anomalies from all solar flux data sets show good agreement with the observations, despite the different flux variations. The off-line CTM reproduces these changes through dynamical information contained in the analyses. A notable feature during this period is a robust positive solar signal in the tropical middle stratosphere, which is due to realistic dynamical changes in our simulations. Ozone changes in the lower mesosphere cannot be used to discriminate between solar flux data sets due to large uncertainties and the short time span of the observations. Overall this study suggests that, in a CTM, the UV variations detected by SORCE are not necessary to reproduce observed stratospheric O3 changes during 2001–2010.

scholarly journals Statistical diagnostic and correction of a chemistry-transport model for the prediction of total column ozone

2006 ◽  
Vol 6 (2) ◽  
pp. 525-537 ◽  
Author(s):  
S. Guillas ◽  
G. C. Tiao ◽  
D. J. Wuebbles ◽  
A. Zubrow
Keyword(s):  
Transport Model ◽  
Chemical Transport ◽  
Satellite System ◽  
Chemical Transport Model ◽  
Quasi Biennial Oscillation ◽  
Data Set ◽  
Total Column Ozone ◽  
Noise Component ◽  
Column Ozone ◽  
Total Column

Abstract. In this paper, we introduce a statistical method for examining and adjusting chemical-transport models. We illustrate the findings with total column ozone predictions, based on the University of Illinois at Urbana-Champaign 2-D (UIUC 2-D) chemical-transport model of the global atmosphere. We propose a general diagnostic procedure for the model outputs in total ozone over the latitudes ranging from 60° South to 60° North to see if the model captures some typical patterns in the data. The method proceeds in two steps to avoid possible collinearity issues. First, we regress the measurements given by a cohesive data set from the SBUV(/2) satellite system on the model outputs with an autoregressive noise component. Second, we regress the residuals of this first regression on the solar flux, the annual cycle, the Antarctic or Arctic Oscillation, and the Quasi Biennial Oscillation. If the coefficients from this second regression are statistically significant, then they mean that the model did not simulate properly the pattern associated with these factors. Systematic anomalies of the model are identified using data from 1979 to 1995, and statistically corrected afterwards. The 1996–2003 validation sample confirms that the combined approach yields better predictions than the direct UIUC 2-D outputs.

Sign in / Sign up

Export Citation Format

Share Document