Figures & Tables (gcb-figures-and-tables)

SkillMedia

Use when building figures, tables, and the mandatory graphical abstract for a Global Change Biology (GCB) manuscript. GCB rewards mechanism-first, self-contained, accessible exhibits and requires a graphical abstract depicting the driver-to-response link. Guides exhibit design; it does not invent data.

Instructions available. Your AI can read the instructions. Execution depends on the setup they require.

Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.

Then ask your AI: use the Figures & Tables (gcb-figures-and-tables) skill

What this skill tells your AI

The instructions your AI receives, as published by brycewang-stanford/awesome-journal-skills in Global-Change-Biology-Skills/skills/gcb-figures-and-tables/SKILL.md and read by ahel’s review.

In GCB, exhibits carry the mechanism. The strongest papers show a clear driver → biological response relationship in the main figures, with uncertainty visible. GCB also requires a graphical abstract that depicts the mechanistic link (not a site map or phylogeny). Verify current figure specs on the live guidelines page before submission.

When to trigger

  • Designing the main-figure sequence to tell the mechanistic story
  • Building the graphical abstract
  • Preparing maps, flux/time-series, dose-response, or forest plots
  • Checking accessibility, resolution, and self-containment before submission

What GCB rewards

  1. Mechanism-first main figures. Lead with the driver-response relationship; relegate site maps and descriptive context to supporting figures unless they carry the argument.
  2. Visible uncertainty. Confidence/credible ribbons, error bars, ensemble spread — uncertainty is part of the result, not optional decoration.
  3. A strong graphical abstract. One concise figure linking an environmental driver to a biological response; it is the reader's first impression.
  4. Self-contained exhibits. Each figure/table readable from its caption alone (units, n, sample, statistic, scale).
  5. Accessibility. Colour-blind-safe palettes (viridis, Okabe-Ito), legible in grayscale; vector line art; live-check the current resolution and file-format specifications.

Tables

  • Report effect sizes with intervals and units; define every symbol and abbreviation in the caption.
  • Keep tables for precise numbers; move relationships and patterns into figures.

Exhibit-by-exhibit expectations

Different exhibit types carry different burdens at GCB. Use this as a per-panel design contract before you finalize the figure set.

ExhibitCarriesMust showCommon reject note
Graphical abstractThe driver → response linkA causal arrow from environmental driver to biological responseA site map or phylogeny standing in for mechanism
Flux / time-series panelTrend and its uncertaintyConfidence ribbon and the sampling unitA bare line with no spread
MapSpatial patternColour-blind-safe ramp, scale bar, projectionRainbow ramp, no scale, undefined extent
Dose-responseMechanism shapeFitted curve with interval and the dataCurve drawn through points with no band
Forest plot (synthesis)Pooled effect + heterogeneityPer-study and pooled effect with CIsPooled diamond only, studies hidden

Worked micro-example (illustrative)

A remote-sensing study of carbon flux wants one mechanistic main figure. A site map of the eddy-covariance towers is supporting material, not the lead. The lead panel plots gross primary productivity against growing-season temperature with a fitted ribbon: GPP rises then saturates near an illustrative 22 C, with the ensemble spread shaded. The caption states n = 18 site-years (illustrative), the statistic, the units (g C m^-2 d^-1), and reads alone. The graphical abstract distils this to a single arrow: warming → productivity saturation → weakening land carbon sink. Numbers illustrative.

Reviewer pushback on exhibits and the fix

  • "Figure 1 is a map; where is the mechanism?" → promote the driver-response panel to lead; demote the map to supporting information.
  • "Point estimates only" → add ribbons, error bars, or ensemble spread so uncertainty is visible.
  • "Unreadable in grayscale / rainbow map" → switch to viridis or Okabe-Ito and verify grayscale legibility.
  • "Caption not self-contained" → add n, units, statistic, and scale so the exhibit stands without the methods.

Anti-patterns

  • A graphical abstract that is a study-site map or phylogeny instead of the driver-response mechanism
  • Main figures that describe the site rather than show the mechanism
  • Plots with no uncertainty (point estimates only)
  • Rainbow/jet colour maps; figures unreadable in grayscale or by colour-blind readers
  • Captions that require the methods section to interpret the figure

Output format

【Graphical abstract】shows driver → response mechanism? [Y/N]
【Main-figure story】mechanism-first sequence? [Y/N]
【Uncertainty shown】ribbons / bars / ensemble spread? [Y/N]
【Self-contained】each caption stands alone (units, n, stat)? [Y/N]
【Accessible】colour-blind-safe + grayscale-legible + resolution? [Y/N]
【Next】gcb-reporting-and-data-policy

Supplementary resources

Signals

GitHub stars
1k
Forks
155
Last commit
Sep 2026
Advanced
Item type
skill
Key
gcb-figures-and-tables
Source
github.com/brycewang-stanford/awesome-journal-skills