Sato and Ise (submitted) Open Data

______________________________________________________<br> Sato and Ise (submitted) Open Data Author: Hisashi SATO (JAMSTEC) hsatoscb_(at)_gmail.com<br> Date: 3 Dec 2020<br> URLs:<br> https://ebcrpa.jamstec.go.jp/~hsato/Sato_and_Ise_2020<br> ______________________________________________________ 1. Folder "1.VCDs"<br> It contains Visualized Climate Environments (VCEs) for training and testing the Convolutional-Neural-Network (CNN) model. Names of compressed files (*.tar.gz) correspond to experiments. Each compressed file contains 4 to 16 folders. Naming rule of folders:<br> The strings "CRU", "NCEP", "HadGEM", and "Miroc" stand for that the files are made from CRU_TS4.0, NCEP/NCAR reanalysis, Had2GEM-ESM, and Miroc-ESM climate datasets, respectively. The strings "AMeans" and "MMeans", respectively, stand for that annual-mean and monthly-mean climates are represented by VCEs in the compressed file.<br> The string "EachYear" means that the compressed files contains VCEs of each year climate from 1971 to 1980, otherwise the compressed files contains VCEs of averaged climate over 10 years. The strings "hist", "RCP26", and "RCP85" mean that the compressed files contain VCEs of climate averaged over 1971-1980, 2091-2100@RCP2.6, and 2091-2100@RCP8.5, respectively. MainSimulation_training.tar.gz<br> VCEs for training the CNN model for the main simulation and dependency test of climatic datasets for training and reconstructing performances. MainSimulation.tar.gz<br> VCEs for testing the CNN model for the main simulation. CombinationSelection_Amean.tar.gz<br> VCEs for an experiment to find optimal combination of climatic variables (annual means) represented by the VCEs. In the VCE of the RGB color tile, up to three climate variables can be represented by RGB channels. To find the optimal combination of climatic variables, we systematically evaluated the model performance of 14 combinations of climatic variable experiments for annual means. It's result is presented in the Supplemental Material 2.<br> Extracting this compressed fire results in "CRU", "NCEP", "HadGEM", and "Miroc" folders. Each folder contains 14 subfolders named with sequential numbers, which corresponds to model numbers in the Supplemental Material 2. CombinationSelection_Mmean.tar.gz<br> Same as the CombinationSelection_Amean.tar.gz except made with monthly mean climates, and related information is available in the Supplemental Material 3. ScalerSelection.tar.gz<br> VCEs for evaluate the influences of different transformations of climatic variables on the resulting accuracy. It's result is presented in the Supplemental Material 4.<br> Extracting this compressed fire results in "CRU", "NCEP", "HadGEM", and "Miroc" folders. Each folder contains 4 subfolders named with sequential numbers. "Scaler1" stands for log transformed, "Scaler2" stands for no transformed (linear), "Scaler3" stands for Sigmoid(gain=5) transformed, "Scaler4" stands for Sigmoid(gain=10) transformed. ColorAssignExperiment.tar.gz<br> VCEs for evaluate the influences of assignment patterns of air temperature and precipitation to RGB color channels of the VCE. It's result is presented in the Supplemental Material 5. Extracting this compressed fire results in "CRU", "NCEP", "HadGEM", and "Miroc" folders. Each folder contains 6 subfolders named with sequential numbers, which indicates model number in the Supplemental Material 5. AverageExtentExperiment.tar.gz<br> VCEs for sensitivity test, where training and test accuracies are compared among models those are trained by monthly climate averaged over three different periods: 10 years (1971-1980; Control), 20 years (1961-1980), and 30 years (1951-1980). Individual VCE shows climatic condition of each half degree grid cell. According to the ISLSCP2 data, VCEs are classified by their their potential vegetation type of the grid. Number of deepest folder names correspond vegetation code of the ISLSCP2. Following are the vegetation code.<br> 01: Tropical Evergreen Forest/Woodland<br> 02: Tropical Deciduous Forest/Woodland<br> 03: Temperate Broadleaf Evergreen Forest/ Woodland<br> 04: Temperate Needleleaf Evergreen Forest/Woodland<br> 05: Temperate Deciduous Forest/Woodland<br> 06: Boreal Evergreen Forest/Woodland<br> 07: Boreal Deciduous Forest/Woodland<br> 08: Evergreen/Deciduous Mixed Forest<br> 09: Savanna<br> 10: Grassland/Steppe<br> 11: Dense Shrubland<br> 12: Open Shrubland<br> 13: Tundra<br> 14: Desert<br> 15: Polar Desert/Rock/Ice Naming rule for VCEs of 10 years average climate is following:<br> Latitude number + "_" + Longitude number + ".png" Naming rule for VCEs of each year climate is following:<br> Latitude number + "_" + Longitude number + "_" + Year of climate + ".png" Here, latitude number ranges from 001 to 360, starting from north-latitude-90 to southward with half degree interval. The longitude number ranges from 001 to 720, starting from west-longitude-180 to eastward with half degree interval. On the top of this subfolder, there is PicList.zip, which is a compressed file of PicList.txt. This text file is <br> required for classify VCEs with the trained model. _____________________________________________<br> 2. Folder "2.Result"<br> It contains copies of image classifications results of the Digits screen. <br> Subfolder names correspond to experiment names. Each subfolder contains subsub-folders "DigitsOutput", "ConfusionMatrix", and "ReconstructedBiomeMap". Naming rules for files are basically same as those of VCEs. _____________________________________________<br> 3. Folder "3.LearningCurves"<br> It contains screen capture of Digits output, showing learning curve of CNN models. CRU climate data. This folder contains 5 subfolders. Naming rules of these subfolders are same as those of VCEs. _____________________________________________<br> 4. Folder "5.FigureMaterials" <br> It contains data and codes (in R) for drawing the figures in the manuscript" _____________________________________________________________佐藤与伊势（已投稿）开放数据集 作者：佐藤 久志（日本海洋科技中心，JAMSTEC） 邮箱：hsatoscb_(at)_gmail.com 日期：2020年12月3日 数据链接：https://ebcrpa.jamstec.go.jp/~hsato/Sato_and_Ise_2020 1. 文件夹「1.VCDs」 该文件夹包含用于训练和测试卷积神经网络（Convolutional-Neural-Network, CNN）模型的可视化气候环境（Visualized Climate Environments, VCEs）。压缩文件（*.tar.gz）的名称对应不同实验，每个压缩文件包含4至16个子文件夹。文件夹命名规则如下： 字符串「CRU」、「NCEP」、「HadGEM」、「Miroc」分别代表数据集源自CRU_TS4.0、NCEP/NCAR再分析数据、Had2GEM-ESM及Miroc-ESM气候数据集。「AMeans」和「MMeans」分别代表该压缩文件中的VCEs对应年平均气候与月平均气候。 「EachYear」代表该压缩文件包含1971至1980年逐年气候的VCEs；反之，压缩文件包含10年平均气候的VCEs。「hist」、「RCP26」、「RCP85」分别代表压缩文件包含1971-1980年平均气候、2091-2100年RCP2.6情景下平均气候、2091-2100年RCP8.5情景下平均气候的VCEs。 MainSimulation_training.tar.gz：用于训练主模拟所用CNN模型的VCEs，同时可用于训练数据集的依赖性测试与性能重构测试。 MainSimulation.tar.gz：用于测试主模拟所用CNN模型的VCEs。 CombinationSelection_Amean.tar.gz：用于探寻以VCEs表征的气候变量最优组合（年平均）的实验所需VCEs。在RGB色块VCE中，最多可通过RGB三个通道表征三种气候变量。为寻找最优气候变量组合，本研究系统评估了14组年平均气候变量组合实验的模型性能，相关结果见补充材料2。解压该压缩文件后将得到「CRU」、「NCEP」、「HadGEM」及「Miroc」四个文件夹，每个文件夹包含14个以连续数字命名的子文件夹，对应补充材料2中的模型编号。 CombinationSelection_Mmean.tar.gz：与CombinationSelection_Amean.tar.gz内容一致，仅基于月平均气候生成，相关信息见补充材料3。 ScalerSelection.tar.gz：用于评估气候变量的不同变换方式对模型最终精度的影响的VCEs，相关结果见补充材料4。解压该压缩文件后将得到「CRU」、「NCEP」、「HadGEM」及「Miroc」四个文件夹，每个文件夹包含4个以连续数字命名的子文件夹：「Scaler1」代表对数变换，「Scaler2」代表无变换（线性变换），「Scaler3」代表Sigmoid（增益=5）变换，「Scaler4」代表Sigmoid（增益=10）变换。 ColorAssignExperiment.tar.gz：用于评估气温与降水分配至VCE的RGB颜色通道的不同模式对模型的影响的VCEs，相关结果见补充材料5。解压该压缩文件后将得到「CRU」、「NCEP」、「HadGEM」及「Miroc」四个文件夹，每个文件夹包含6个以连续数字命名的子文件夹，对应补充材料5中的模型编号。 AverageExtentExperiment.tar.gz：用于敏感性测试的VCEs，该测试对比了使用三种不同时段月平均气候训练的模型的训练与测试精度：10年（1971-1980；对照组）、20年（1961-1980）及30年（1951-1980）的月平均气候。每个VCE对应一个半度网格单元的气候状况。根据ISLSCP2数据，VCEs按网格单元的潜在植被类型进行分类，最深层文件夹的名称对应ISLSCP2的植被代码，植被代码如下： 01：热带常绿林/林地 02：热带落叶林/林地 03：温带阔叶常绿林/林地 04：温带针叶常绿林/林地 05：温带落叶林/林地 06：北方常绿林/林地 07：北方落叶林/林地 08：常绿/落叶混交林 09：稀树草原 10：草原/干草原 11：密灌丛 12：开阔灌丛 13：苔原 14：荒漠 15：极地荒漠/岩石/冰原 10年平均气候的VCEs命名规则如下：纬度编号_经度编号.png 逐年气候的VCEs命名规则如下：纬度编号_经度编号_气候年份.png 其中，纬度编号范围为001至360，从北纬90度向南以半度间隔递增；经度编号范围为001至720，从西经180度向东以半度间隔递增。该子文件夹根目录下存在PicList.zip，内含PicList.txt压缩文件，该文本文件用于使用训练好的模型对VCEs进行分类。 2. 文件夹「2.Result」 该文件夹包含Digits可视化界面的图像分类结果副本。子文件夹名称对应实验名称，每个子文件夹包含三个子子文件夹：「DigitsOutput」、「ConfusionMatrix」及「ReconstructedBiomeMap」。文件命名规则与VCEs的命名规则基本一致。 3. 文件夹「3.LearningCurves」 该文件夹包含Digits输出的屏幕截图，展示了CNN模型的学习曲线，对应CRU气候数据集。本文件夹包含5个子文件夹，其命名规则与VCEs的命名规则一致。 4. 文件夹「5.FigureMaterials」 该文件夹包含绘制论文配图所用的数据及R语言代码。