Skip to content

Input Data

Required inputs

The function canyonb requires seven input parameters for prediction:

Parameter Description Units Range Typical Error
Time Measurement date UTC or decimal year Any -
Latitude Geographic latitude degrees -90 to 90 -
Longitude Geographic longitude degrees -180 to 180 -
Pressure Water pressure dbar 0 to 10000 0.5
Temperature In-situ temperature °C -2 to 35 0.005
Salinity Practical salinity PSU 0 to 45 0.005
Oxygen Dissolved oxygen μmol·kg−1 0 to 400 1% of value

Data Format

Array Requirements

All input arrays must:

  • Have the same shape
  • Be numpy arrays or convertible to numpy arrays
  • Can contain non-finite values (np.nan), but not recommanded for coordinates (time, longitude, latitude, pressure)
import numpy as np
from datetime import datetime

# Single point
data = {
    'gtime': [datetime(2024, 1, 1)],
    'lat': np.array([45.0]),
    'lon': np.array([-20.0]),
    'pres': np.array([100.0]),
    'temp': np.array([15.0]),
    'psal': np.array([35.0]),
    'doxy': np.array([250.0])
}

# Multiple points/profile
depths = np.array([0, 100, 200, 500])
data_profile = {
    'gtime': [datetime(2024, 1, 1)] * len(depths),
    'lat': np.full_like(depths, 45.0),
    'lon': np.full_like(depths, -20.0),
    'pres': depths,
    'temp': np.array([20.0, 15.0, 12.0, 8.0]),
    'psal': np.full_like(depths, 35.0),
    'doxy': np.array([250.0, 220.0, 200.0, 180.0])
}

Time Format

CANYON-B accepts two time formats:

  1. Python datetime objects:

    from datetime import datetime
    time = [datetime(2024, 1, 1)]
    

  2. Decimal years:

    time = [2024.0]  # January 1st, 2024
    

Geographic Coordinates

Latitude

  • Must be between -90° and 90°
  • Negative values for Southern hemisphere and positive values for Northern hemisphere

Longitude

Can be in either (Automatically converted internally):

  • -180° to 180° format (preferred)
  • 0° to 360° format

Arctic Region Handling

For measurements in the Arctic basin west of the Lomonosov ridge:

  • Latitude values are automatically adjusted
  • Adjustments account for water mass distribution
  • No user action required