March 2015
Spotlight Summary by Anne Hirsikko
Retrieval of temperature from a multiple-channel Rayleigh-scatter lidar using an optimal estimation method
Temperature is one of the fundamental variables describing the state of the atmosphere, and the ability to observe temperature changes reliably is essential for applications ranging from climate research to weather forecasting. Life on the ground is directly influenced by the atmosphere and weather in the lower atmosphere. However, the state of the middle atmosphere (approximately at 15-90 km altitude), and chemical and dynamical processes therein have an indirect influence on the ground through interactions between atmospheric layers. Temperature observations are frequently made with in-situ sensors near surface. Additionally, temperature profiles through the atmosphere are continuously being measured with passive and active remote sensors from the ground and satellite-borne. Rayleigh-backscatter lidar, which is based on elastic backscatter of light by objects much smaller than the wavelength, is one of the most commonly deployed sensing techniques. Due to technical limitations (one sensor measures over a certain specific altitude range), though, it is best to use a combination of sensors to characterize temperature through the atmosphere.
Previous temperature retrieval methods applied to Rayleigh-scatter lidar observations suffer from shortcomings and limitations, notably their incomplete uncertainty characterization. The paper by Sica and Haefele introduces important advances in the retrieval of temperature profiles from Rayleigh-scatter lidar by applying the optimal estimation method (OEM). Earlier, OEM has been successfully used with passive sensors for obtaining information about the atmospheric state, or with elastic backscatter lidars for analyzing aerosol properties.
OEM requires a forward model which best describes the system. In their paper, Sica and Haefele present performance comparisons for two forward models, which were chosen following careful investigations. These models are based essentially on the lidar equation with or without the assumption of hydrostatic equilibrium. Both forward models showed good performance and comparability against modelled temperature profiles when using synthetically produced lidar photocount profiles. However, once measured lidar photocount profiles were used, the forward model with released assumption of hydrostatic equilibrium produced different temperature profiles when compared to other methods. This discrepancy was shown to arise from the pressure profiles being used, which must necessarily be accurate. Therefore, with currently available pressure profiles, the forward operator that assumes hydrostatic equilibrium is better at producing the desired temperature profiles.
A typical problem when data from two lidar channels are combined is the merging of observations, and subsequently of their uncertainties, right in an overlapping region. The method applied by Sica and Haefele leads to a smooth transition between two observation profiles. Thus, one of the main benefits of OEM is its potential for synergetic temperature retrieval from multiple observation sources. Yet another advantage is that this method allows improved budgeting of uncertainties that are related to retrieved temperature profiles.
As a conclusion, OEM showed very good applicability and significant improvements in temperature retrievals from multi-channel Rayleigh backscatter lidar observations. Therefore, it is expected that this method will become more widely used among the lidar community.
You must log in to add comments.
Previous temperature retrieval methods applied to Rayleigh-scatter lidar observations suffer from shortcomings and limitations, notably their incomplete uncertainty characterization. The paper by Sica and Haefele introduces important advances in the retrieval of temperature profiles from Rayleigh-scatter lidar by applying the optimal estimation method (OEM). Earlier, OEM has been successfully used with passive sensors for obtaining information about the atmospheric state, or with elastic backscatter lidars for analyzing aerosol properties.
OEM requires a forward model which best describes the system. In their paper, Sica and Haefele present performance comparisons for two forward models, which were chosen following careful investigations. These models are based essentially on the lidar equation with or without the assumption of hydrostatic equilibrium. Both forward models showed good performance and comparability against modelled temperature profiles when using synthetically produced lidar photocount profiles. However, once measured lidar photocount profiles were used, the forward model with released assumption of hydrostatic equilibrium produced different temperature profiles when compared to other methods. This discrepancy was shown to arise from the pressure profiles being used, which must necessarily be accurate. Therefore, with currently available pressure profiles, the forward operator that assumes hydrostatic equilibrium is better at producing the desired temperature profiles.
A typical problem when data from two lidar channels are combined is the merging of observations, and subsequently of their uncertainties, right in an overlapping region. The method applied by Sica and Haefele leads to a smooth transition between two observation profiles. Thus, one of the main benefits of OEM is its potential for synergetic temperature retrieval from multiple observation sources. Yet another advantage is that this method allows improved budgeting of uncertainties that are related to retrieved temperature profiles.
As a conclusion, OEM showed very good applicability and significant improvements in temperature retrievals from multi-channel Rayleigh backscatter lidar observations. Therefore, it is expected that this method will become more widely used among the lidar community.
Add Comment
You must log in to add comments.
Article Information
Retrieval of temperature from a multiple-channel Rayleigh-scatter lidar using an optimal estimation method
R. J. Sica and A. Haefele
Appl. Opt. 54(8) 1872-1889 (2015) View: Abstract | HTML | PDF