COBE2全球海表温度和海冰分析资料（1850-2019）

COBE 百年海温和海冰分析数据是日本气象厅研制的一套空间分布完整的海温分析数据。该数据集融合了ICOADS2.5、船测和浮标报告、全球电信系统（Global Telecommunications System, GTS）资料，经最优插值到1°×1°经纬网格上得到。在1941年之前，本数据集使用桶测数据进行误差订正。在进行内插分析之前，数据均经过先验阈值的质量控制，同时，附近的观测也被综合考虑。本数据集有助于对过去百年全球气候变化的理解，也可为大气环流模式提供质量较好的边界条件。JRA再分析数据和AMIP试验均使用了COBE数据集。观测数据的质量控制基于检验观测报告中的船轨迹、日期、位置，不正确的数据根据日本气象厅搜集的海洋气象资料被自动订正，以当日观测为中心点的三个月之内，数据误差数量过大的船舶信号会被拉入黑名单。在本数据集当中，海冰－海温关系表达式得到了改进，该表达式用来通过海冰观测资料来估计海温（南北极海洋）。此外，在本数据集的客观分析之前，每种类型的海温观测的全球平均海温均一化序列的误差进行了估计，这些误差的大小用来确定海温资料和夜间气温观测资料的一致性，经过误差订正的全球平均海温序列与已有研究结果一致。在观测稀疏地区，卫星观测也参加了数据重建。另外，对区域平均海温序列的理论分析误差也进行了考察的同时，还进行了样本误差估计。涉及变量包括：逐月海表温度、逐月海冰密集度、逐月海表温度和海冰气候平均（1981—2010年平均）。

COBE Centennial Sea Surface Temperature and Sea Ice Analysis Dataset is a fully spatially resolved sea surface temperature (SST) analysis dataset developed by the Japan Meteorological Agency (JMA). It integrates data from ICOADS2.5, shipboard and buoy reports, and Global Telecommunications System (GTS) observations, and is optimally interpolated onto a 1°×1° latitude-longitude grid. Prior to 1941, error correction was performed using bucket measurement data for this dataset. Before conducting interpolation analysis, all data underwent quality control with prior thresholds, and nearby observations were comprehensively considered.

This dataset facilitates understanding of global climate change over the past century, and can also provide high-quality boundary conditions for atmospheric circulation models. Both the JRA reanalysis dataset and the AMIP experiment have utilized the COBE dataset.

The quality control of observational data is based on the inspection of ship trajectories, dates, and positions in observation reports. Incorrect data are automatically corrected based on marine meteorological data collected by the JMA. Ship signals with an excessive number of data errors within a three-month period centered on the observation day will be blacklisted.

In this dataset, the sea ice-SST relationship expression has been improved, which is used to estimate SST in the Arctic and Antarctic oceans using sea ice observation data. Additionally, prior to the objective analysis of this dataset, errors in the globally averaged SST homogenized sequences for each type of SST observation were estimated. The magnitude of these errors was used to determine the consistency between SST data and nighttime air temperature observation data. The error-corrected globally averaged SST sequence is consistent with existing research results.

In regions with sparse observational coverage, satellite observations are also employed for data reconstruction. In addition, theoretical analysis errors of regional averaged SST sequences were investigated, and sample error estimation was also conducted.

The variables involved include: monthly mean sea surface temperature, monthly mean sea ice concentration, and the climatological mean of monthly sea surface temperature and sea ice (average for the period 1981–2010).

搜集汇总

背景与挑战

背景概述

COBE2数据集是由日本气象厅研制的全球海表温度和海冰分析资料，覆盖1850年至2019年，时间分辨率为月，空间分辨率为1°×1°经纬网格。该数据集融合了多种观测资料（如ICOADS2.5、船测和浮标报告），经过最优插值和严格的质量控制（包括误差订正和改进的海冰－海温关系），提供逐月海表温度、海冰密集度及气候平均变量，主要用于全球气候变化研究和大气环流模式边界条件。