Dataset for Deep Tillage Impacts on Soil Properties and Crop Productivity

Data Description for Deep Tillage Impacts on Soil Properties and Crop Productivity This dataset is associated with a systematic review aiming to assess the global impacts of deep tillage on soil properties and crop productivity. All data were extracted from peer-reviewed literature published up to October 2025, following strict inclusion criteria and standardized data extraction procedures, to ensure the reliability and comparability of the data for subsequent meta-analysis and related research. 1. Data Source Relevant studies were retrieved from four major academic databases to ensure comprehensive coverage of global research findings: ISI Web of Science, Google Scholar, SpringerLink, and the China Knowledge Resource Integrated Database. The search strategy employed a combination of key terms related to deep tillage, including “sub-soiling”, “subsoiling”, “deep till*”, “deep plough*”, “deep rip*”, “deep mixing”, and their related combinations. No restrictions were imposed on the publication year except for the cut-off date of October 2025. 2. Study Eligibility and Screening Eligible studies were selected based on five predefined inclusion criteria to ensure the validity and relevance of the data: Language Requirement: Publications must be written in English or Chinese; Chinese-language articles were restricted to high-quality journals indexed in the Chinese Science Citation Database (CSCD) to guarantee academic rigor. Experimental Type: Only field-based experiments were included; pot trials, greenhouse studies, and laboratory simulations were excluded, as they cannot fully reflect the actual field conditions of deep tillage application. Comparative Design: Studies must report at least one direct paired comparison between deep tillage and conventional tillage treatments, with a minimum of three replicates per treatment to ensure statistical robustness. Location Documentation: Experimental locations must be clearly recorded, either through GPS coordinates or explicit city-level identification, to facilitate spatial analysis of deep tillage impacts. Consistent Management: Other agricultural management practices (including straw return methods, irrigation regimes, and fertiliser application rates and types) were required to remain consistent between deep tillage and conventional tillage treatments, eliminating confounding factors that could affect soil properties and crop productivity. After a rigorous screening and selection process, 298 publications met all the inclusion criteria and were included in the dataset. 3. Data Content and Structure The dataset contains a total of 3,114 crop yield observations collected from 357 experimental sites worldwide (see associated Fig. 1 for spatial distribution). For each crop yield observation, the following key information was extracted and organized: Basic study information: Author(s), publication year, journal name, experimental location (GPS coordinates or city-level location). Tillage treatment details: Specific types of deep tillage and conventional tillage, tillage depth, and tillage frequency. Crop-related information: Crop type, planting density, growth period, and management practices (straw return, irrigation, fertilisation) consistent across treatments. Yield data: Mean crop yield, standard deviation (SD), and number of replicates for both deep tillage and conventional tillage treatments. 4. Data Extraction and Quality Control Data extraction was conducted using standardized procedures to ensure accuracy and consistency: Data Extraction Tools: Tabulated data were directly extracted from the literature; graphical data (e.g., bar charts, line graphs) were digitized using GetData Graph Digitizer (version 2.26) to obtain mean values and associated variability. Standard Deviation (SD) Calculation: Where studies only reported standard errors (SE), SD values were calculated using the formula$$SD = SE \times \sqrt{n}$$, where $$n$$ is the number of replicates. SD Estimation: In cases where neither SD nor SE was reported, SD was estimated using Bracken’s method implemented in the metagear package (version 4.5.2) in R software, ensuring that variability information was available for all observations. Quality Check: All extracted data were cross-checked by two independent researchers; discrepancies were resolved through re-examination of the original literature and consensus discussion to minimize extraction errors. 5. Data Usage Notes This dataset is suitable for meta-analyses, spatial pattern analyses, and comparative studies focusing on the global impacts of deep tillage on crop productivity. Users are advised to: Refer to the original publications for detailed experimental design and additional context not included in this dataset. Consider the potential influence of regional climate, soil type, and crop species when interpreting the data, as these factors may modulate the effects of deep tillage. Cite both this dataset (via its assigned DOI) and the original studies when using the data in academic research or publications. The dataset is provided in a structured format (e.g., CSV) for easy access and analysis, with clear column headers and detailed annotations to facilitate understanding and usage.