Abstract
Abstract. Clouds present many challenges to climate modelling. To develop and verify the parameterisations needed to allow climate models to represent cloud structure and processes, there is a need for high-quality observations of cloud optical depth from locations around the world. Retrievals of cloud optical depth are obtainable from radiances measured by Aerosol Robotic Network (AERONET) radiometers in “cloud mode” using a two-wavelength retrieval method. However, the method is unable to detect cloud phase, and hence assumes that all of the cloud in a profile is liquid. This assumption has the potential to introduce errors into long-term statistics of retrieved optical depth for clouds that also contain ice. Using a set of idealised cloud profiles we find that, for optical depths above 20, the fractional error in retrieved optical depth is a linear function of the fraction of the optical depth that is due to the presence of ice cloud (“ice fraction”). Clouds that are entirely ice have positive errors with magnitudes of the order of 55 % to 70 %. We derive a simple linear equation that can be used as a correction at AERONET sites where ice fraction can be independently estimated. Using this linear equation, we estimate the magnitude of the error for a set of cloud profiles from five sites of the Atmospheric Radiation Measurement programme. The dataset contains separate retrievals of ice and liquid retrievals; hence ice fraction can be estimated. The magnitude of the error at each location was related to the relative frequencies of occurrence in thick frontal cloud at the mid-latitude sites and of deep convection at the tropical sites – that is, of deep cloud containing both ice and liquid particles. The long-term mean optical depth error at the five locations spans the range 2–4, which we show to be small enough to allow calculation of top-of-atmosphere flux to within 10 % and surface flux to about 15 %.
Highlights
Clouds are a crucial part of the climate system, yet present many great challenges to climate science (Randall et al, 2007; Boucher et al, 2013)
We propose a simple linear correction equation that could be employed in Aerosol Robotic Network (AERONET) locations where ice fraction can be independently determined
The analysis above from the five Atmospheric Radiation Measurement (ARM) sites implies that, if an estimate of ice fraction is not available at a given AERONET site, using uncorrected retrieved optical depths will lead to a mean error of the order of 2–4 in long-term statistics
Summary
Clouds are a crucial part of the climate system, yet present many great challenges to climate science (Randall et al, 2007; Boucher et al, 2013). Climate models struggle to represent the optical properties of clouds (Bender et al, 2006; Lauer and Hamilton, 2013; Klein et al, 2013; Calisto et al, 2014). Cloud optical depth is important to represent reliably as it governs the effect of clouds on the Earth’s radiation budget. The complex processes and interactions that describe the evolution of clouds occur on scales much smaller than a model grid box and require parameterisation (Pincus et al, 2003; Shonk and Hogan, 2010). To develop and validate these parameterisations, there is a need for global observations of cloud optical depth at high temporal and spatial resolution. A common approach to measure cloud optical depth is to retrieve it remotely from measurements of reflectance, radi-
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call