关键词: 主粮作物, 棉油作物, 生育状况, 资料整合, 质量控制

Abstract:

A dataset of the growth conditions of major crops in China was mainly constructed from paper-based annual records before 2012 and electronic annual records after 2013. However, there were problems such as inconsistencies in the observed items and data units，the quality of some data had not been evaluated. To improve the consistency and accuracy of agricultural meteorological data, based on these two types data, a high-quality dataset of the growth conditions China's major crops (including wheat, rice, maize, cotton, oil-seed rape, soybean and peanut) from 1981 to 2022 was developed by using the observation items standardization, integrity checks, cross-year value checks, observation time checks, value range checks, internal consistency checks element limit value check and manual verification. The dataset promoted effective application in agricultural research and decision-making. The results showed that the valid rate of crop common stage from 1981 to 2022 was over 96.0% of the expected observations, while the valid rate for growth status, crop height, stem count and effective stem count were all over 86.0%. The accuracy rate of the above five mentioned elements were above 99.3%. The distribution of observation stations for the seven major crops had obvious spatial and temporal distribution characteristics, with dense stations, uniform spatial distribution and long observation years in eastern China, but sparse and short observation years in northwest China. There were also obvious differences in the number of observation stations between different crops, and the number of observation stations for cotton and oil crops were less than that for staple crops. The valid data was relatively low in the 1980s, but improved significantly after 1994. After quality control and data verification, the valid rate of crop common stage increased from 94.7% to 96.2%, the crop height increased from 88.2% to 92.0%, the stem count increased from 77.1% to 86.7%. The accuracy rate of the common stage data increased from 99.3% to 99.6%. Compared to the "China Major Crops Growth and Development Dataset V1.0"，the overall quality of this dataset has been improved, with the addition of element boundary value checks. This dataset can provide critical fundamental information for studying the impact of climate change on the growth and development of major crops in China.

Key words: Staple crops, Cotton and oil-seed crops, Growth and development, Data merge, Quality control