climplot

Publication-quality climate science plotting utilities for Python.

Mission: Help climate science beginners make beautiful, publication-ready plots with minimal effort.

Features

• Style Modes: Switch between publication and presentation styles with one function call
• Climate Colormaps: Discrete colormaps optimized for anomaly fields, with center-on-white option
• Map Utilities: Easy map creation with Cartopy projections
• Multi-panel Figures: Consistent panel labeling and colorbars
• Area-weighted Metrics: Accurate statistics for gridded climate data

Installation

pip install climplot

Or install from source:

git clone https://github.com/jkrasting/climplot.git
cd climplot
pip install -e .

Quick Start

import climplot
import matplotlib.pyplot as plt

# Set publication style
climplot.publication()

# Create a map figure
fig, ax = climplot.map_figure()

# Plot with discrete colormap
cmap, norm, levels = climplot.anomaly_cmap(-0.3, 0.3, 0.05)
cs = ax.pcolormesh(lon, lat, data, cmap=cmap, norm=norm, transform=ccrs.PlateCarree())

# Add colorbar
climplot.add_colorbar(cs, ax, 'SSH Anomaly (m)')

# Save
climplot.save_figure('my_figure.png')
plt.close()

Style Modes

Publication Mode

For journal figures with small, dense typography:

• 3.5" width (single-column), 7.0" (two-column)
• 8-11pt fonts
• 300 DPI

climplot.publication()  # Single column
climplot.publication(width=7.0)  # Two-column
climplot.publication(for_pdf=True)  # PDF with embedded fonts

Presentation Mode

For slides and posters with larger, readable typography:

• 7.0" width
• 12-16pt fonts
• 150 DPI

climplot.presentation()
climplot.presentation(for_pdf=True)  # PDF for slides

Colormaps

Anomaly Colormap

Red-blue diverging, centered on zero:

cmap, norm, levels = climplot.anomaly_cmap(-0.3, 0.3, 0.05)

Center-on-White

For difference plots where near-zero values should appear neutral:

cmap, norm, levels = climplot.anomaly_cmap(-0.3, 0.3, 0.05, center_on_white=True)

Auto-Levels

When you don't know the data range in advance, auto_levels picks a nice interval automatically:

interval, levels = climplot.auto_levels(-2.7, 2.7, n_levels=10)
cmap, norm, _ = climplot.discrete_cmap(levels[0], levels[-1], interval)

Log-Scale Colormap

For data spanning orders of magnitude (e.g., tracer concentrations):

cmap, norm, levels = climplot.log_cmap(0.01, 100, per_decade=3)

Colorbar Customization

add_colorbar supports keyword arguments for tick density and sizing:

climplot.add_colorbar(cs, ax, 'Precip (mm/day)', max_ticks=7, width=0.03)

Maps

Projections

# Robinson projection (Pacific-centered)
fig, ax = climplot.map_figure()

# Atlantic-centered
fig, ax = climplot.map_figure(central_longitude=0)

# Different projection
fig, ax = climplot.map_figure(projection='mollweide')

Rendering Land: Three Workflows

Atmosphere / lat-lon grids (reanalysis, CMIP atmos, observations):

fig, ax = climplot.map_figure()
cmap, norm, levels = climplot.anomaly_cmap(-2, 2, 0.5)
cs = climplot.plot_atmos_field(
    ax, lon, lat, temperature,
    cmap=cmap, norm=norm, levels=levels,
)
climplot.add_gridlines(ax, x_spacing=30, y_spacing=30)  # optional
climplot.add_colorbar(cs, ax, 'Temperature Anomaly (K)')

plot_atmos_field draws light-gray land underneath, plots data with slight transparency on top, and adds thin coastlines — all in one call.

Regular / regridded grids (observations, reanalysis on 1°×1°, etc.):

fig, ax = climplot.map_figure()
cs = ax.pcolormesh(lon, lat, data, cmap=cmap, norm=norm,
                   transform=ccrs.PlateCarree())
climplot.add_land_feature(ax)    # filled gray continents from Natural Earth
climplot.add_coastlines(ax)      # optional coastline outlines

Native ocean-model grids (tripolar, MOM6 — Cartopy coastlines won't align):

fig, ax = climplot.map_figure()
cs = climplot.plot_ocean_field(
    ax, geolon_c, geolat_c, sst,
    wet_mask=wet,   # 1=ocean, 0=land
)

plot_ocean_field sets a gray background so NaN over land renders as the model's coastline, optionally masks land, and plots the data in one call. See the Plotting Guide for full details.

Multi-panel Figures

# 2x3 panel figure
fig, axes = climplot.panel_figure(2, 3)

# Add panel labels (a. b. c. etc.)
climplot.add_panel_labels(axes.flatten())

# Single colorbar below all panels
climplot.bottom_colorbar(cs, fig, axes, 'Temperature (K)')

Area-weighted Metrics

Why area weighting matters: Simple averaging over-weights polar regions. This can produce errors of ~1 cm in global sea level means.

import climplot

# Area-weighted mean (CORRECT)
gmsl = climplot.area_weighted_mean(ssh, areacello, dim=['yh', 'xh'])

# Simple mean (WRONG - ~1 cm error)
# gmsl = ssh.mean(dim=['yh', 'xh'])

# Other metrics
bias = climplot.area_weighted_bias(model, obs, areacello, dim=['yh', 'xh'])
rmse = climplot.area_weighted_rmse(model, obs, areacello, dim=['yh', 'xh'])
corr = climplot.area_weighted_corr(model, obs, areacello, dim=['yh', 'xh'])

# Comprehensive summary
metrics = climplot.metrics_summary(model, obs, areacello, dim=['yh', 'xh'])
climplot.print_metrics_summary(metrics, name='My Model')

Dependencies

• matplotlib >= 3.7
• numpy >= 1.24
• xarray >= 2023.1
• cartopy >= 0.21

License

MIT License - see LICENSE

Contributing

Contributions are welcome! Please see our contributing guide.