Evaluating Arctic cloud radiative effects simulated by NICAM with A‐train. Issue 12 (18th June 2016)

• Record Type: Journal Article
• Authors: Hashino, Tempei; Satoh, Masaki; Hagihara, Yuichiro; Kato, Seiji; Kubota, Takuji; Matsui, Toshihisa; Nasuno, Tomoe; Okamoto, Hajime; Sekiguchi, Miho
• Is Part Of: Journal of geophysical research. Volume 121:Issue 12(2016)
• Page Start: 7041
• Page End: 7063
• Publication Date: 2016-06-18
• Subjects: model evaluation -- A‐train -- global cloud‐resolving model -- satellite data simulator -- cloud radiative effects -- cloud microphysics; Atmospheric physics -- Periodicals; Geophysics -- Periodicals; 551.5
• DOI: 10.1002/2016JD024775
• Languages: English
• ISSNs: 2169-897X

Abstract

Abstract: Evaluation of cloud radiative effects (CREs) in global atmospheric models is of vital importance to reduce uncertainties in weather forecasting and future climate projection. In this paper, we describe an effective way to evaluate CREs from a 3.5 km mesh global nonhydrostatic model by comparing it against A‐train satellite data. The model is the Nonhydrostatic Icosahedral Atmospheric Model (NICAM), and its output is run through a satellite‐sensor simulator (Joint Simulator for satellite sensors) to produce the equivalent CloudSat radar, CALIPSO lidar, and Aqua Clouds and the Earth's Radiant Energy System (CERES) data. These simulated observations are then compared to real observations from the satellites. We focus on the Arctic, which is a region experiencing rapid climate change over various surface types. The NICAM simulation significantly overestimates the shortwave CREs at top of atmosphere and surface as large as 24 W m −2 for the month of June. The CREs were decomposed into cloud fractions and footprint CREs of cloud types that are defined based on the CloudSat‐CALIPSO cloud top temperature and maximum radar reflectivity. It turned out that the simulation underestimates the cloud fraction and optical thickness of mixed‐phase clouds due to predicting too little supercooled liquid and predicting overly large snow particles with too little mass content. This bias was partially offset by predicting too many optically thin high clouds. Offline sensitivity experiments, where cloud microphysical parameters, surface albedo, and single scattering parameters are varied, support the diagnosis. Aerosol radiative effects and nonspherical single scattering of ice particles should be introduced into the NICAM broadband calculation for further improvement. Key Points: Cloud albedo/shading and greenhouse effects of NICAM were underestimated compared with A‐train data The biases can be connected to cloud microphysics and cloud types Too frequent and too thin high clouds partially compensate for lack of supercooled water clouds