Data and analysis code for three-year field responses of orange and lemon to soil-applied copper and zinc
Description
Data and analysis code underlying a three-year field study of soil-applied copper and zinc in orange and lemon on acidic yellow-gleyic soil in Lankaran, Azerbaijan. The package contains archived soil-profile measurements, annual yield summaries for 2021-2023, 2023 fruit-quality and fruit-mineral values, 2023 replicate-level yield data, and Python code reproducing the conditional exact permutation and Page tests. Design limitations and the discrepancy between replicate-derived and institute-reported 2023 means are documented explicitly.
Files
Steps to reproduce
1. Install Python 3 (tested with Python 3.12) and the pinned dependencies: pip install -r requirements.txt (NumPy 2.3.5, SciPy 1.17.0) 2. From inside the extracted package folder (so the script finds the data file), run: python Citrus_Cu_Zn_analysis.py 3. The script reads yield_2023_replicates.csv and performs the conditional 2023 analysis: an exact omnibus permutation test on the additive-block treatment F statistic (complete enumeration of 1,728,000 within-block arrangements) and an exact one-sided Page test for the ordered dose response (216 arrangements per series, Holm-adjusted across the four series). 4. Expected output, matching the manuscript: Orange: treatment F = 61.23, omnibus exact p = 0.000069 Lemon: treatment F = 12.86, omnibus exact p = 0.000417 Page test: L = 42 (maximum), exact one-sided p = 0.0046 per series Holm-adjusted Page p = 0.019 across the four series 5. The remaining CSV files (soil_profile_2020, annual_yield_2021_2023, fruit_quality_orange_2023, fruit_quality_lemon_2023) contain the descriptive data reported in Tables 1-5 for transparency and reuse; they are not required to run the script. Yields are in centner ha-1 (1 centner ha-1 = 0.1 t ha-1); the permutation and Page results are scale-invariant.