Changes in cropland soil inorganic carbon and its relationship with nitrogen fertilization and precipitation over the past 40 years in the Sichuan Basin, China

Widespread soil acidification driven by nitrogen (N) fertilization and precipitation challenges the conventional notion of the long-term stability of soil inorganic carbon (SIC) in agroecosystems. However, the changes in SIC with precipitation and N fertilization remain ambiguous. Based on 4,000+ soil samples collected in the 1980s and 2010s and by developing machine learning models to fill the missing SIC of soil samples, this study generated 3,697 paired soil samples between the two periods and then investigated the cropland SIC change and explored its relationship with precipitation and N fertilization across the Sichuan Basin, China. The results showed an overall SIC loss, with a decline of the mean SIC by 15.73%. SIC change varied with initial soil pH and initial SIC and exhibited an exponential relationship with soil pH change, indicating the changing role of carbonates in providing acid-buffering capacity. There was a parabolical relationship between the magnitude of SIC decline and N fertilizer rates, and low N fertilizer rates contributed to a reduction in SIC loss, while SIC loss was promoted by N fertilization occurred when N fertilizing rates exceeded 250 kg ha^–1 yr^–1. The change in SIC showed a sinusoidal variation with precipitation, with 950 mm being the threshold controlling whether SIC increased or decreased. Meanwhile, N fertilization did not alter the sinusoidal relationship between SIC change and precipitation. In areas with rainfall<950 mm, the high N fertilizer rate did not cause SIC loss, while higher precipitation could also cause larger SIC loss in areas with lower N fertilizer rates. These results suggest that SIC dynamics are jointly driven by precipitation and N fertilization and are controlled by acid-buffering mechanisms associated with initial pH and SIC, with precipitation being the predominant driver. These findings emphasize the need for more regional soil observations and in-depth studies of SIC change and its mechanisms for accurately estimating SIC change.