The continuous expansion of non-agricultural transformation and non-grain production of arable land has seriously affected China's food security, and its evolution mechanism and control measures have already become an important subject. Given more existing researches focusing on the one-side discussion of non-agricultural transformation or non-grain production on a small scale, it is necessary to carry out the quantitative assessment research on the interaction among non-agricultural transformation, non-grain production and rural depopulation on the spatiotemporal perspective. Based on four periods of land use data and agricultural labor data in 1990, 2000, 2010, and 2020, the spatio-temporal evolution characteristics of non-agricultural transformation, non-grain production, and rural population hollowing in county level of the Yangtze River Economic Belt were analyzed. The coupling and coordination degree between non-agricultural transformation, non-grain production and rural population hollowing was calculated using coupling and coordination analysis. Then, the spatio-temporal dynamics of the coupling and coordination degree were analyzed based on exploratory spatio-temporal data analysis (ESTDA). The results showed that: (1) There were significant disparities in the indicators of non-agricultural transformation, non- rain production and rural population hollowing. The non-agricultural transformation presented a pattern of high value in the east and low value in the west, and there was a trend of high value clustering around the central city. Non-grain production generally presented a tendency of west high and east low. The degree of non-grain production was higher in the remote areas far from big cities. During the study period, the trend of non-agricultural and non-grain production was intensified, and the hollowing degree of the rural population increased significantly. (2) There was a strong coupling effect between non-agricultural transformation and rural population hollowing, and the imbalance area was gradually expanding. (3) The coupling coordination degree of non-agricultural, non-grain and rural population hollowing had obviously spatial agglomeration phenomenon and gradually increases. The high-value clusters were mainly distributed in the upstream area and the number was gradually decreasing, while the low-value clusters were relatively dispersed. (4) The local spatial pattern of coupling coordination degree of non-agricultural, non-grain and rural population hollowing had strong spatio-temporal dynamics, and showed an increasing trend from the western region to the eastern region. The changing process of local dependencies was relatively stable which indicates the high development inertia of local dependencies. The proportion of cooperative change between counties and its neighborhood was high, indicating strongly local integration.