OLCI script AMT paper

Analysis code for "Boundary layer turbulence characterization from OLCI-derived integrated water vapor structure functions: a global multi-site analysis" (Kermarrec, 2026, AMT).

Contents: olci_amt_analysis.m (main script, reproduces all figures and tables), seasonal_count.m (monthly/seasonal product counts).

Requirements: MATLAB R2022b+, Statistics and Machine Learning Toolbox, Mapping Toolbox (optional for world map).

Input data: (1) OLCI SF results as .mat files with fields heff_ref (L₀), a1_ref (α₁), a2_ref (α₂), C_ref (σ²), sf_ref (60-bin SF), centers, heff_detection_rate, iwv_mean, produced by olci_batch_v4b_generic.m. (2) ERA5 .nc files: hourly BLH (era5_{site}_hourly_*.nc) and sub-daily wind/TCWV (era5_{site}_single_*.nc) from the Copernicus Climate Data Store.

Usage: Set MAT_DIR, ERA5_DIR, ERA5_HOURLY_DIR at the top of olci_amt_analysis.m, then run. Figures are saved to ./figures/. Run seasonal_count.m afterwards for product distribution tables.

Sites: Highveld (n=42), India (n=115, 2016–2024), Pampas (n=58), US Plains (n=177, 2016–2024), Yangtze (n=31).

Key parameters: RD_MIN=0.85, HEFF_MIN/MAX=350/3000 m, MAD_THRESH=3.5, SMOOTH_WIN=5 months.

Contact: Gaël Kermarrec, kermarrec@meteo.uni-hannover.de, Institute of Meteorology and Climatology, Leibniz Universität Hannover.