地表入射太阳辐射是地表能量的主要来源，对大气环流、水循环和生态系统等具有重要影响。近些年来，随着卫星技术的不断发展和应用，使用遥感反演方法获取全球范围高时空分辨率的地表太阳辐射通量已经成为研究地表能量平衡的重要数据来源之一。因此，卫星太阳辐射数据的质量能够直接影响地表能量平衡的研究。以往卫星太阳辐射的评估工作主要集中在月尺度上的比较，对于更精细的小时尺度评估工作相对较少。有研究表明CERES(Clouds and the Earth's Radiant Energy System)卫星月数据精度明显高于大部分再分析数据和其他卫星数据，然而其在小时尺度的表现如何还不明确。据此，本研究利用BSRN(Baseline Surface Radiation Network，BSRN)地面观测网络中的38个站点近十年的小时观测数据来定量评估CERES SYN1deg-1Hour数据的精度。所得到的主要结论有：

（1）CERES卫星小时太阳辐射数据在全球范围内整体上的偏差较小，在大多数站点整体全年偏差在10W/m2之内。卫星高估的站点略多于低估的站点，少部分地理位置较特殊的站点存在较大偏差，偏差大的站点多为负偏差；位于极圈内的站点数据质量尤其差。

（2）在日内尺度上，卫星太阳辐射在大部分站点在辐射较为强烈的正午标准化偏差与均方根误差略低，相关系数较高，而在辐射弱的傍晚则是相关系数较低，标准化偏差与均方根误差较高。在北美洲区域的站点，数据质量较高，日内的波动较小；在高纬度地区站点数据质量较差，日内各地方时的各项统计指标波动较大。

（3）在季节尺度上，卫星太阳辐射在不同月份间的标准化偏差、均方根误差以及相关系数有着较为明显的季节变化规律。多数站点在辐射较强的夏季标准化偏差与均方根误差较小，相关系数较高，而在冬季的月份则相反。卫星太阳辐射在北半球站点的各个统计指标的季节变化比南半球的季节变化更加明显。北美洲区域卫星太阳辐射数据质量较好，极圈内的站点则质量较差。

（4）纬度是影响站点间偏差分布状况的重要因素，高纬站点的观测质量明显较低纬站点更差。此外，卫星太阳辐射数据在月份上的变化较日内地方时的变化更为明显，这说明影响卫星数据质量的主要是不同季节条件下气候态的差异而并非太阳高度角的变化。

The surface incident solar radiation is the main source of surface energy, which has an important impact on the atmospheric circulation, water cycle and ecosystem. In recent years, with the continuous development and application of satellite technology, using remote sensing inversion method to obtain global high spatial and temporal resolution of surface solar radiation flux has become one of the important data sources to study the surface energy balance. Therefore, the quality of satellite solar radiation data can directly affect the study of surface energy balance. In the past, the satellite solar radiation assessment mainly focused on the comparison on the monthly scale, but less on the more precise hourly scale. Some studies have shown that the monthly data accuracy of CERES(Clouds and the Earth's Radiation Energy System) satellite is significantly higher than that of most reanalysis data and other satellite data, but its performance in the hourly scale is not clear. Therefore, the accuracy of CERES SYN1deg-1 hour data was quantitatively evaluated by using the hourly observation data of 38 bsrn stations in the past ten years. The main conclusions are as follows：

(1) The global deviation of CERES hourly solar radiation data is small, and the annual deviation is within 10W/m2 in most stations. The overestimated sites are slightly more than the underestimated sites, a small number of sites with special geographical location have large deviation, and the sites with large deviation are mostly negative deviation; The data quality of stations located in the polar circle is particularly poor.

(2) On the intraday scale, the standard deviation and root mean square error of satellite solar radiation at most stations are slightly lower and the correlation coefficient is higher at noon when the radiation is relatively strong, while the correlation coefficient is lower and the standard deviation and root mean square error are higher at dusk when the radiation is weak. In the North American region, the data quality is high, and the intra day fluctuation is small; In the high latitude area, the data quality of the station is poor, and the statistical indicators of each place and time fluctuate greatly.

(3) On the seasonal scale, the standard deviation, root mean square error and correlation coefficient of satellite solar radiation in different months have obvious seasonal variation. The results show that the standard deviation and root mean square error of most stations are small in summer with strong radiation, and the correlation coefficient is high, while the opposite is true in winter. The seasonal variation of each statistical index of solar radiation in the northern hemisphere is more obvious than that in the southern hemisphere. The quality of satellite solar radiation data in North America is better than that in polar regions.

(4) Latitude is an important factor affecting the distribution of inter station deviation. The observation quality of high latitude stations is worse than that of low latitude stations. In addition, the variation of satellite deviation in month is more obvious than that in local time, which indicates that the difference of climate state in different seasons rather than the change of solar altitude angle is the main factor affecting the quality of satellite observation.