A dataset of the impacts of no-till and reduced-till on soil organic carbon and crop yield in black soil region of Northeast China

All extracted data were standardized in terms of units and format—for example, SOC content was expressed as g C kg⁻¹, and crop yields as t ha⁻¹. To ensure data completeness, annual mean temperature (MAT) and annual precipitation (MAP) values not explicitly reported were estimated using the WorldClim database and the latitude and longitude of the study sites. The moisture index (HI) was calculated as MAP divided by annual evapotranspiration (MAE: HI = MAP/MAE).

Missing values for initial soil properties such as pH, bulk density (BD), initial SOC content, total nitrogen (TN), C/N ratio, and the proportions of sand, silt, and clay particles were obtained from the Harmonized World Soil Database v1.2 (HWSD). Throughout the data processing process, key parameters were manually entered by team members and carefully double-checked to minimize errors and ensure accuracy. Normality tests were also conducted on continuous variables like SOC and crop yields.

The dataset covers the period from 1983 to 2025, spanning 43 years and providing a basis for studying the long-term effects of no-till/reduced-till practices. Geographically, the data pertains to the Black Soil Region of Northeast China, located between 122°44′–132°48′E longitude and 40°28′–50°15′N latitude. The dataset is stored in a single Excel file (.xlsx format, 1.85 MB) and contains 838 rows of observational data, with each row representing a comparison between no-till/reduced-till treatments and conventional tillage controls.

The data includes basic information about the studies, location details, tillage treatments, response variables (SOC and crop yields), and physical, chemical, and biological properties of the soil after treatment. In the dataset, CK and T represent the mean and standard deviation of the control and treatment groups, respectively; N denotes the sample size, and Unit indicates the unit of measurement. Some statistical indicators and explanatory variables have missing values, primarily due to differences in the level of detail provided in the original studies.

To improve data completeness, missing values were filled in and marked in red in the dataset to alert users to their supplementary nature during analysis. Errors in the data stem mainly from potential inaccuracies in data extraction from publications and estimation errors related to imputed missing values. To minimize these errors, the research team conducted thorough double-checks and ensured that all data was standardized.