A compressed NumPy version of the EMAG2 (v3) global Earth Magnetic anomaly grid compiled from satellite, ship, and airborne magnetic measurements. The original CSV data was imported, transformed, and saved to a compressed NumPy archive as follows: import numpy as np mag_data = np.loadtxt('EMAG2_V3_20170530.csv', delimiter=',', usecols=(2,3,4,5,7)) lon_mask = mag_data[:,0] > 180.0 mag_data[lon_mask,0] -= 360.0 np.savez_compressed('EMAG2_V3_20170530.npz', data=mag_data.astype(np.float32)) The NumPy archive (contained in this repository) can be efficiently loaded in Python workflows. It contains the following columns: Longitude - geographic longitudinal coordinates in decimal degrees (WGS84) Latitude - geographic latitudinal coordinates in decimal degrees (WGS84) SeaLevel - magnetic anomaly value at sea level (nT) UpCont - magnetic anomaly value at continuous 4km altitude (nT) Error - Error estimate (nT) Code 888 is assigned in certain cells on grid edges where the data source is ambiguous and assigned an error of -888 nT. Code 999 is assigned in cells where no data is reported with the anomaly value assigned 99999 nT and an error of -999 nT. Reference Brian Meyer, Richard Saltus, Arnaud Chulliat (2017): EMAG2: Earth Magnetic Anomaly Grid (2-arc-minute resolution) Version 3. National Centers for Environmental Information, NOAA. Model. doi:10.7289/V5H70CVX