Toward autonomous surface-based infrared remote sensing of polar clouds: cloud-height retrievals

Penny M Rowe,Von P Walden,Christopher J Cox

doi:10.5194/amt-9-3641-2016

Abstract

Abstract. Polar regions are characterized by their remoteness, making measurements challenging, but an improved knowledge of clouds and radiation is necessary to understand polar climate change. Infrared radiance spectrometers can operate continuously from the surface and have low power requirements relative to active sensors. Here we explore the feasibility of retrieving cloud height with an infrared spectrometer that would be designed for use in remote polar locations. Using a wide variety of simulated spectra of mixed-phase polar clouds at varying instrument resolutions, retrieval accuracy is explored using the CO2 slicing/sorting and the minimum local emissivity variance (MLEV) methods. In the absence of imposed errors and for clouds with optical depths greater than ∼ 0.3, cloud-height retrievals from simulated spectra using CO2 slicing/sorting and MLEV are found to have roughly equivalent high accuracies: at an instrument resolution of 0.5 cm−1, mean biases are found to be ∼ 0.2 km for clouds with bases below 2 and −0.2 km for higher clouds. Accuracy is found to decrease with coarsening resolution and become worse overall for MLEV than for CO2 slicing/sorting; however, the two methods have differing sensitivity to different sources of error, suggesting an approach that combines them. For expected errors in the atmospheric state as well as both instrument noise and bias of 0.2 mW/(m2 sr cm−1), at a resolution of 4 cm−1, average retrieval errors are found to be less than ∼ 0.5 km for cloud bases within 1 km of the surface, increasing to ∼ 1.5 km at 4 km. This sensitivity indicates that a portable, surface-based infrared radiance spectrometer could provide an important complement in remote locations to satellite-based measurements, for which retrievals of low-level cloud are challenging.

Highlights

Measurements of cloud properties are needed to improve climate and forecast models of the Arctic and Antarctic atmospheres (Hines et al, 2004; Town et al, 2007; Wesslen et al, 2014)
Methods for combining minimum local emissivity variance (MLEV) and CO2 slicing/sorting are worth pursuing but are beyond the scope of this work. They could include combining them at the algorithmic level: for example, in a Bayesian analysis that determines the optimal solution based on the intersection of the mean ±1 standard deviation probabilities for CO2 slicing/sorting and MLEV
The extra computational time taken for running both CO2 slicing/sorting and MLEV is minimal because the most time-consuming computations are the calculations of Bc, tc, and Rc for each model layer, and this set of calculations is identical for the two methods

Summary

Introduction

Measurements of cloud properties are needed to improve climate and forecast models of the Arctic and Antarctic atmospheres (Hines et al, 2004; Town et al, 2007; Wesslen et al, 2014). Measurements of cloud properties at high latitudes come primarily from satellite platforms (e.g., Wang and Key, 2005; Lubin et al, 2015). Active instruments, such as lidar, can vertically profile clouds (see, e.g., Verlinden et al, 2011; Cesana et al, 2012) but have a small footprint, so that monthly or seasonal averaging is needed for global coverage. Passive instruments that measure upwelling infrared radiances have large footprints, enabling global coverage on daily timescales

Methods

Results

Discussion

Conclusion