Vertical distribution of the particle phase in tropical deep convective clouds as derived from cloud-side reflected solar radiation measurements

Evelyn Jäkel, Manfred Wendisch, Trismono Candra Krisna, Florian Ewald, Tobias Kölling, Tina Jurkat, Christiane Voigt, Micael A. Cecchini, Luiz A. T. Machado, Armin Afchine, Anja Costa, Martina Krämer, Meinrat O. Andreae, Ulrich Pöschl, Daniel Rosenfeld, Tianle YuanAtmospheric 2017 · DOI

TL;DR

This study uses airborne spectral measurements of deep convective clouds (DCCs) to map the vertical distribution of their particle phase (liquid water vs. ice). They found that polluted conditions tend to push the mixed-phase layer higher, and that cloud development stage significantly impacts this vertical structure, with implications for understanding cloud radiative effects and the hydrological cycle.

What Problem Does This Paper Address?

Deep convective clouds (DCCs) are crucial for Earth's climate system, playing a major role in heat redistribution and the hydrological cycle. Understanding their internal structure, particularly the transition from liquid water to ice (the mixed-phase layer), is vital. This transition zone significantly influences cloud properties like reflectivity and lifetime. However, accurately measuring this vertical phase distribution, especially in the complex environment of tropical DCCs, has been challenging. Previous methods often relied on satellite observations of cloud tops or in-situ measurements that provide only point data, making it difficult to capture the full vertical profile and its variability. This paper aims to address this by using a novel cloud-side remote sensing approach to derive detailed vertical profiles of the particle phase in tropical DCCs and investigate how aerosol conditions and cloud development stages influence this structure.

Methods

The core of this research is a novel retrieval method for determining the particle phase of DCCs using airborne measurements. The key techniques include:

  • Cloud-Side Spectral Reflectivity Measurements: The study utilizes data from the specMACS (spectrometer of the Munich Aerosol Cloud Scanner) imaging spectroradiometer. This instrument captures spectral information from the sides of clouds, avoiding the complexities of penetrating thick clouds.
  • Phase Discrimination Retrieval: A phase index ($I_p$) is calculated based on the differential absorption of solar radiation by ice and liquid water particles in the near-infrared spectrum (specifically between 1550 and 1700 nm). This index is sensitive to the particle phase, with negative values indicating liquid water and positive values indicating ice.
  • Cloud Masking: To ensure accurate measurements, a method was developed to distinguish between illuminated and shadowed cloud regions. This is crucial because shadowed regions are affected by radiation from other parts of the cloud, contaminating the spectral signal. The method uses a simplified RGB color value derived from spectral radiances to identify illuminated areas.
  • Cloud Geometry Retrieval: To translate spectral measurements into vertical profiles, the distance to the cloud and its height are needed. This is achieved using stereophotogrammetric techniques with collocated images from a GoPro camera. By comparing images taken from slightly different aircraft positions, 3D cloud geometry can be reconstructed.
  • Radiative Transfer Modeling: A 3D radiative transfer model (MCARATS) was used to simulate spectral radiances and validate the phase index retrieval method under various cloud conditions and viewing geometries.

Datasets

The study primarily uses data collected during the ACRIDICON-CHUVA campaign over the Brazilian rainforest in September 2014. Specifically, data from three scientific flights (AC10, AC13, and AC18) were analyzed.

In addition to the airborne measurements, the study also uses:

  • MODIS (Moderate-resolution Imaging Spectroradiometer) cloud products: These were used for comparison with the derived phase profiles and for applying an ensemble method to estimate glaciation temperature.
  • In situ measurements: Data from instruments like NIXE-CAPS (Novel Ice eXpEriment – Cloud, Aerosol and Precipitation Spectrometer) and CAS-DPOL (Cloud and Aerosol Spectrometer) were used to validate the remote sensing results.

Key Insights and Ideas

  • Aerosol Influence on Mixed-Phase Layer Height: Polluted aerosol conditions appear to push the mixed-phase layer higher in DCCs compared to moderate aerosol conditions. This suggests that aerosols can influence cloud microphysics and radiative properties by altering the altitude at which ice formation begins.
  • Cloud Development Stage Matters: The vertical distribution and depth of the mixed-phase layer are highly variable, even within a single cloud scene. Mature or decaying clouds, affected by falling ice particles, show lower levels of fully glaciated cloud layers compared to actively growing clouds. This highlights the dynamic nature of DCCs.
  • Cloud-Side Remote Sensing as a Powerful Tool: The study demonstrates the effectiveness of using cloud-side spectral reflectivity measurements, combined with stereophotogrammetry for geometry retrieval, to obtain detailed vertical profiles of cloud particle phase. This approach overcomes limitations of both satellite (cloud-top only) and in-situ (point measurements) methods.
  • Variability Within Cloud Clusters: Even clouds in close proximity can exhibit different phase states at the same altitude, suggesting that localized processes like updrafts and downdrafts, and precipitation, play a significant role in shaping the vertical phase structure.

Key Results

  • The depth and vertical position of the mixed-phase layer can vary by up to 900 m within a single cloud scene.
  • For polluted aerosol conditions (AC13), the lower boundary of the mixed-phase layer was found at 6.2 ± 0.3 km (267 K), and the upper boundary at 7.4 ± 0.4 km (259 K).
  • For moderate aerosol conditions (AC18), the lower boundary was at 5.6 ± 0.2 km (269 K), and the upper boundary at 6.8 ± 0.2 km (263 K). This confirms the theoretical expectation of a higher mixed-phase layer in polluted conditions.
  • In situ measurements of liquid water content (LWC) and asphericity (AF) from NIXE-CAPS and CAS-DPOL generally agreed with the remote sensing-derived mixed-phase zone, extending from 6.4 to 8.7 km for AC13.

Limitations and Future Work

  • Ensemble Method Limitations: The study notes that the satellite-based ensemble method, while useful for large cloud fields, may not accurately represent individual clouds, especially when phase distribution is altered by updrafts and downdrafts.
  • Temporal and Spatial Variability: Direct comparisons between different observation strategies (cloud-side, cloud-top, and in situ) are subject to restrictions due to the inherent temporal and spatial variability of convective systems.
  • Future Work: The authors plan to conduct future studies focusing on observing individual convective clouds to document their entire evolution from growing to dissipating stages. They also intend to utilize the ATTO (Amazonian Tall Tower Observatory) for continuous monitoring, providing a valuable platform for studying DCCs in a highly variable environment.

Summary updated

Abstract

Vertical profiles of cloud particle phase in tropical deep convective clouds (DCCs) were investigated using airborne solar spectral radiation data collected by the German High Altitude and Long Range Research Aircraft (HALO) during the ACRIDICON-CHUVA campaign, which was conducted over the Brazilian rainforest in September 2014. A phase discrimination retrieval based on imaging spectroradiometer measurements of DCC side spectral reflectivity was applied to clouds formed in different aerosol conditions. From the retrieval results the height of the mixed-phase layer of the DCCs was determined. The retrieved profiles were compared with in situ measurements and satellite observations. It was found that the depth and vertical position of the mixed-phase layer can vary up to 900 m for one single cloud scene. This variability is attributed to the different stages of cloud development in a scene. Clouds of mature or decaying stage are affected by falling ice particles resulting in lower levels of fully glaciated cloud layers compared to growing clouds. Comparing polluted and moderate aerosol conditions revealed a shift of the lower boundary of the mixed-phase layer from 5.6 ± 0.2 km (269 K; moderate) to 6.2 ± 0.3 km (267 K; polluted), and of the upper boundary from 6.8 ± 0.2 km (263 K; moderate) to 7.4 ± 0.4 km (259 K; polluted), as would be expected from theory.
BibTeX
@inproceedings{jakel2017vertical,
  author = {Evelyn Jäkel and Manfred Wendisch and Trismono Candra Krisna and Florian Ewald and Tobias Kölling and Tina Jurkat and Christiane Voigt and Micael A. Cecchini and Luiz A. T. Machado and Armin Afchine and Anja Costa and Martina Krämer and Meinrat O. Andreae and Ulrich Pöschl and Daniel Rosenfeld and Tianle Yuan},
  title = {{Vertical distribution of the particle phase in tropical deep convective clouds as derived from cloud-side reflected solar radiation measurements}},
  booktitle = {Atmospheric},
  year = {2017},
  doi = {10.5194/acp-17-9049-2017},
}
Figure