N-P-K fertilizer application rates by crop and country for circa 2020
Data files
Jul 31, 2025 version files 10.57 MB
-
circa2020_grids.zip
9.95 MB
-
README.md
2.77 KB
-
Supplementary_Data_1.xlsx
615.16 KB
Abstract
All fertilizer application rates are on elemental basis (P = P2O5 x 0.436; K = K2O x 0.830). Imputation to build complete crop x year N-P-K time series data: Subnational- to global-scale understanding is built using FUBC information collected from individual countries and sub-nationally with ~800,000 data points for 1980-2022. Briefly, a three-step imputation process with an additional preprocessing was used. In the first step long-term average FUBC was determined. In the second step the exact year observations was identified, which is then allowed to pass on unchanged. In a final third step the numbers from the first and second step are readjusted (observations are not adjusted) to flexibly match the total agricultural fertilizer consumption in a country as reported by the FAO. Sub-national FUBC rates for ten major crops in 25 agriculturally important countries use a similar three-step iteration method. For the remaining countries, only the national FUBC values are reported. For mapping the fertilizer application rates of the ten major crops an annual gridded crop dataset was used.
Dataset DOI: 10.5061/dryad.59zw3r2mq
Description of the data and file structure
Description of the data present in each sheet of the Excel workbook (Supplementary_Data_1.xlsx):
The data is present in several sheets.
Sheet 1 is named “Description”, and it contains metadata explaining the Excel data file structure.
Sheet 2 is named “Regional Summary” and it contains information on N-P-K fertilizer application rates for circa 2020 at global and regions scales for all crops and specific crops. Column A in this “Regional Summary” sheet is the name of the region. Column B is the data period, which is always circa 2020 (2018-2022 average). Column C in the “Regional Summary” sheet is the FAO denoted crop code, and column D is the crop name. For rows 4 to 23, we provide a summary for all crops, which is exactly what it means - fertilizer applications on average, for all crops. The application rates are given in columns E-G (N, P, and K-fertilizer application rates, respectively), followed by consumption (in metric tons) columns H-J and harvested hectares in column K (in thousands of hectares) in the “Regional Summary” sheet.
This is followed by 18 sheets of these specific regions and N-P-K fertilizer application rates for each country of the region for circa 2020, and each crop. These sheets are called: East Africa, East Asia, North America, South America, South Asia, Western & Middle Africa, Western Europe, Caribbean, Central America, Eastern Europe, North Africa, Northern Europe, Oceania, Southeast Asia, Southern Africa, Southern Europe, and Western Asia.
The column headers in these 18 sheets are Year, Country Code (FAO style), Country Name, FAO Crop Code, Crop name, followed by 3 columns of N, P, and K fertilizer application rates (columns F, G, and H) in the units of kg/ha for circa 2020 and then 3 columns of consumption of N, P, and K fertilizers (in units of metric tons - columns I, J, and K). The last column (column L) is the harvested hectares of the crop in units of thousand hectares. Multiplying the application rates in columns F, G, and H with column L will give columns I, J, and K, respectively, as well.
Description of the circa2020_grids (circa2020_grids.zip):
There are 10 crops: barley, cassava, maize, oil palm, rapeseed, rice, sorghum, soybean, sugarcane, and wheat. Each of the 10 crops has three files grouped under N, P, and K application rates in the netCDF format. Each file is gridded data of dimension 4320 x 2160 grid cells (at 5 min spatial resolution and in units of kg/ha/year). The data can be directly imported into a GIS software such as QGIS.
Code/software
All data were developed using MATLAB.