The numbers of extreme weather and climate events in many parts of the world are increasing due to climate change, and this phenomenon will become more frequent and severe. A complete and reliable climate observation series over long-time scale is an important basis for interpreting climate change and evaluating model simulation performance. It is also a reliable observation basis for deeply and systematically detecting the characteristics of global, regional or local extreme climate change and predicting the future trend of climate change. In this paper, some important research achievements are summarized, which have been recently made in construction of long-term climate observation series on global, regional and local scales at home and abroad. Several global surface temperature and precipitation climate datasets and their important values for the science of climate change are analyzed. Moreover, technologies and key scientific problems in constructing climate observation series over long-time scale that need to be improved and solved urgently are also discussed. Results indicate that some progress has been made in construction technology of climate observation series over long-time scale for supporting the development of climate change operations. However, there are still some main problems that need to be solved in view of the urgent need of data supply capacity under new situation, i. e., enriching the types of the observed elements, restoring the integrity of historical data, improving the quality of data products, and refining the spatial scale of the object and the temporal scale of the elements. Therefore, a scientific basis can be provided for further enhancing the application value of precious long-term climate observations in meteorological service guarantee and scientific and technological innovation in China.

long-time series  /  climate observations  /  homogenization  /  interpolation