Data and code for “Trend reversal and resource planning uncertainty in agricultural mineral potassium fertilizer use in China”

Published: 15 October 2026| Version 1 | DOI: 10.17632/yvmt5dnpw7.1
Contributor:

Description

This dataset provides the data, analysis scripts, model outputs, and supporting documentation associated with the study entitled “Trend reversal and resource planning uncertainty in agricultural mineral potassium fertilizer use in China”. The study examines agricultural mineral potassium fertilizer use in mainland China during 2000–2024, using FAOSTAT data expressed in K₂O equivalents. The dataset includes annual mineral potassium fertilizer use records, harvested-area information for 96 selected crop items (2,400 item-year observations), and derived variables used for trend analysis and area–intensity decomposition. The accompanying materials support the analysis of structural trend changes, Bayesian autoregressive forecasting, rolling-origin forecast evaluation, uncertainty quantification, conditional resource-demand scenarios, and asymmetric-loss decision assessment. The archive contains processed research inputs, Python analysis scripts, model outputs, supporting figures, and documentation describing data sources and reproduction procedures. The materials are intended to improve transparency and reproducibility. Mineral fertilizer use is distinguished from total crop potassium uptake and apparent potash consumption. The dataset does not directly measure soil potassium depletion or substitution by secondary potassium resources. Original data sources include FAOSTAT Fertilizers by Nutrient and FAOSTAT Crops and Livestock Products. Users should consult the accompanying documentation and acknowledge the original data providers.

Files

Steps to reproduce

1. Download the accompanying ZIP archive and extract all files while preserving the original directory structure. 2. Consult README_FIRST.md, DATA_SOURCES_AND_RIGHTS.md, and the documentation included in the archive for data definitions, sources, software requirements, and file descriptions. 3. Use the archived FAOSTAT agricultural mineral potassium fertilizer use series for mainland China (2000–2024) and the harvested-area records for 96 selected crop items. 4. Follow the supplied Python scripts and documentation to reproduce the continuous segmented trend analysis, area–intensity decomposition, Bayesian autoregressive forecasts, rolling-origin model comparisons, sensitivity analyses, and asymmetric-loss evaluation. 5. Compare the generated outputs with the archived model results, tables, and figures. Monte Carlo-based estimates may show minor differences depending on random seeds, software versions, and numerical settings. The archive contains the original analysis materials and saved outputs. The full computational workflow has not been independently rerun as part of this repository preparation. Users should consult the included documentation for reproducibility details and limitations.

Institutions

Categories

Agricultural Economics

Licence