Instruments and Methods Optimization of the Acid Catalyst Concentration for Synthesis of Anti‐Cancer Agent Gamavuton‐0 by Using Mathematical and Statistical Software

Joint Scientific Symposium IJJSS 2016 Chiba, 20‐24 November 2016 19 Topic : Image and Signal Processing Cloud Retrieval and Cloud Type Detection from Himawari‐8 Satellite Data Based on the Split Window Algorithm Babag Purbantoro a,b , Jamrud Aminuddin a,c , Naohiro Manago a , Koichi Toyoshima a , Josaphat Tetuko Sri Sumantyo a , and iroaki Kuze a a Center for Environmental Remote Sensing, Chiba University, 1‐33 Yayoi‐Cho, Inage‐Ku, Chiba, 2638522, JAPAN. b Remote Sensing Technology and Data Center‐Indonesian Institute of Aeronautics and Space, Jl. Lapan No. 70, Pasar Rebo, Jakarta, 13710, INDONESIA. c Department of Physics, Faculty of Mathematics and Natural Science, Universitas Jenderal Soedirman, Jl. dr. Suparno 61 Purwokerto, Jawa Tengah, 53123, INDONESIA Abstract Cloud detection and cloud type classification yields basic information indispensable for meteorological as well as climate studies. ere we propose the use of imawari‐8 data. This satellite has been employed for operational use since July , providing spectral data over channels in visible VS , near‐infrared NR and thermal infrared TR spectral regions. Also, high temporal resolution is available for both full disk min and Japanese archipelago . min . First, we test the traditional approach of the split‐ window algorithm using bands and , centered at and m, respectively, for the determination of cloud density. The brightness temperature difference BTD between two bands are derived on the basis of the Planck formula. The resultant threshold temperatures are and K for the discrimination of high, middle and low‐level clouds. The BTD thresholds, on the other hand, are determined to be and K for distinguishing very thin, thin or thick clouds. The verification of cloud classification results will also be discussed. Keywords Cloud Type Retrieval; imawari‐8; Split Window; Calipso.

1. Introduction

The meteorological satellite imawari 8 was launched by Japan Meteorological Agency JMA in October and started the full operation in July . As compared with its predecessor, MTSAT imawari , significant improvements have been made in regard to the temporal resolution as well as the number of spectral bands. The acquisition frequency of imawari‐8 is . min for Japan area and min for the full disk. t has bands in the visible VS , near infrared NR and thermal infrared TR spectral regions. Because of these advanced features, the satellite provides remote sensing capability in quite a wide perspective, in addition to its original purpose of meteorological monitoring. Corresponding author. Babag Purbantoro, ndonesian National nstitute of Aeronautics and Space Tel.:+ 8 8 fax: + 8 . E‐mail address: babaglapan.go.id 20 Based on their altitudes, clouds can roughly be classified into three classes, namely, low‐level km , middle‐level ‐ km and high‐level km clouds. Because of the lapse rate of troposphere ~ . Kkm , the difference in altitudes leads to the difference in the blackbody temperature T b . Thus, a value of T b can be assigned to each pixel of imawari‐8 imagery, depending on the cloud top height and the Planck formula. n terms of cloud classification, the difference in T b between the thermal infrared bands yields useful information. The split window algorithm noue, 8 , and ; amada et al., has widely been used to discriminate various types of clouds. For example, noue proposed the use of the two dimensional plot of the m T b and the brightness temperature difference BTD between and m bands hereafter BTD ‐ to classify cirrus, dense cirrus, cumulonimbus, and cumulus clouds over the tropical ocean noue, 8 . Split window algorithm was particularly suitable for detecting high‐level cloud type, but occasionally misclassification occurred for low‐ and middle‐level cloud types noue, . Also a multi‐spectral technique to infer the cloud properties using BTD 8‐ and BTD ‐ was proposed to separate water cloud and ice cloud Strabala, . n the present work, we apply BTD ‐ and m to detect the cloud type based on the thickness density of the cloud. To help the correct interpretation, lapse rate formula is used to calculate the altitude and verify the cloud type using Calipso data NASA, . n this research, the cloud types indicated in Figure are examined from the analysis of imawari 8 images around Japan. Figure 1. The split window algorithm.

2. Method

n this research, we use the geo‐corrected full disk data of imawari‐8 downloaded from CEReS, Chiba University ftp:hmwr8 gr.cr.chiba‐u.ac.jpgriddedFDV . ere we use the data cropped only in Japan area . ‐ . N and . ‐ . E . n the first step, radiance I is calculated for each pixel from the digital count, and subsequently for band through , albedo A is calculated by multiplying the radiance with the transformation coefficient c’ JMA, : I = Gain Count + Constant, A = c I . The values of gain and constant are given in the header information of imawari‐8 data. Comparing the histograms of bands – taken midday, we decided to choose band centered at . μm for the initial selection of cloudy pixels because of the best capability of distinguishing cloud and non‐cloud coverage Fig. . ere the resulting criterion is . 8 for cloud‐free area and . 8 for cloudy area. n the second step, we calculate the blackbody temperature, T b , using infrared band of imawari‐8 and the following formula JMA, : , where ln .