Abstract: We installed a set of agrometeorological automatic observation system in Tuquan and Zhalaiteqi respectively, Xingan league, Inner Mongolia in 2016, to perform comparison tests on automatic and manual observation of agrometeorological elements of spring maize and rice on their development period, plant density, plant canopy height, growth status, leaf area index and dry matter weight. According to the comparative test data analysis, the results showed that: the automatic observation data of crop growth status and growth period were basically consistent with the manual observation data and could completely replace the manual observation; automatic observation had relatively large errors on plant density, canopy height, leaf area index and dry matter weight, and could not replace manual observation. The automatic and digital agrometeorological observation of crop growth status and growth stage could enhance the objectivity, timeliness and continuity of observation data and provide data support for improving the accuracy of agrometeorological service.