技能 数据科学 农学展示图表报告规范

农学展示图表报告规范

v20260724
fcr-figures-and-tables
本指南详细介绍了农业科学手稿中创建科学严谨、自洽图表(展示图表)的最佳规范。它涵盖了定量数据报告的最佳实践,包括必须包含的误差项(SED/LSD)、正确的图表选择(如AMMI/GGE生物图、响应曲线)以及确保可访问性和可重复性,以满足农学审稿人的要求。
获取技能
273 次下载
概览

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

Exhibits are where an agronomy reviewer checks whether the result is real and general. At FCR every exhibit must be self-contained and quantitatively complete: units, sample/replication, and a measure of error or variability (SE, SED, or LSD) belong on the exhibit itself.

When to trigger

  • Designing the main results table/figure or a key descriptive exhibit
  • Deciding what belongs in the article vs. supplementary material
  • A reviewer found an exhibit unclear, mislabeled, or missing error/units
  • Presenting G×E, response curves, or model evaluation

Principles

  1. Self-contained. A reader should understand each exhibit from its caption, axis/column labels, and footnote alone. State the crop, cultivar(s), environments (sites×seasons), N/replication, units (SI), and what the value is (mean? adjusted mean?).
  2. Show the error. Yield and treatment means need SED or LSD (with α and df) or error bars defined in the caption — never bare means. For curves, show fitted line + CI and the data.
  3. Right exhibit for the question. Use a response curve for quantitative factors (N, water, density); an AMMI/GGE biplot or Finlay–Wilkinson plot for G×E; observed-vs-simulated with the 1:1 line for model evaluation; time series vs. thermal time/phenology with weather overlays for development.
  4. Accessible. Colourblind-safe palettes; legible in grayscale; no chartjunk, no 3D, no needless colour. Vector output (PDF/EPS) for print.
  5. Reproducible & consistent. Numbers match the analysis script and the deposited data; table and figure values are internally consistent and consistent with the text.

Agronomy-specific exhibits

  • Yield-gap / boundary-line plots; nitrogen- and water-response curves with fitted models.
  • AMMI biplots, GGE biplots, Finlay–Wilkinson stability regressions for multi-environment data.
  • Weather (rainfall, temperature, radiation) shown against crop phenology (sowing, anthesis, maturity).
  • Maps where spatial/regional variation is the point; observed-vs-simulated panels for modelling.

Exhibit-selection table (question → exhibit → annotation)

The right exhibit follows from the agronomic question. Pair each with the annotation an FCR reviewer expects.

Question Exhibit Must annotate
Yield vs. N/water/density fitted response curve + points model, SED or CI, units
Genotype ranking AMMI / GGE / Finlay–Wilkinson environments labelled, % variance
Model performance observed-vs-simulated, 1:1 line RMSE, nRMSE, EF, n; validation only
Treatment means by environment adjusted-means table SED/LSD, α, df, replication

Worked exhibit vignette (illustrative)

Illustrative. A first-draft Table 2 for the maize MET lists raw plot means with a/b/c letters across all 5 N rates and no error term — two flags at once: letters on a quantitative dose hide the response shape, and raw means do not match the mixed-model output. The fix is two exhibits: an N response curve per environment with fitted line and SED bar (α = 0.05), plus an adjusted-means table with one SED column — both self-contained and reproducible from the script.

Anti-patterns

  • Means with no SED/LSD, error bars, or units
  • Mean-separation letters on a quantitative dose where a response curve is appropriate
  • Tables that need the prose to be intelligible (not self-contained)
  • Colour-only encoding that fails in grayscale or for colourblind readers
  • Exhibit values that don't match the analysis output or the data deposit

Operating pass for Field Crops Research

Treat this skill as an executable review pass, not a prose hint. First lock the crop system, environment structure, GxE logic, and yield or physiology endpoint; then judge whether the current manuscript answers the venue's real reader: agronomy reviewers who expect field-based, multi-environment evidence and crop-level general significance.

  • Do the pass: Return a claim-evidence-risk ledger rather than a prose-only diagnosis; every recommendation must point to a manuscript location or missing artifact.
  • Return a ledger: give claim / evidence / risk / manuscript location rows, so the next agent can edit rather than rediscover the issue.
  • Sibling guard: compare against Agricultural Systems for whole-system modeling, European Journal of Agronomy for agronomic breadth, Crop Science for cultivar or breeding emphasis; if a sibling owns the contribution, recommend re-routing before polishing format.
  • Submission-ready gate: do not give final advice until the pack's resources/official-source-map.md has been checked for upload-week rules and the manuscript has one concrete fix for the largest venue-specific risk.

Output format

【Main exhibit】what it shows + why this exhibit type
【Self-contained?】caption + labels + crop/cultivar + envs + N + units present? [Y/N]
【Error shown?】SED / LSD / CI with α stated? [Y/N]
【Accessible?】grayscale-legible + colourblind-safe? [Y/N]
【Article vs supplement】split decided
【Reproducible?】matches analysis output + data deposit? [Y/N]
【Next】fcr-reporting-and-data-policy

Supplementary resources

信息
Category 数据科学
Name fcr-figures-and-tables
版本 v20260724
大小 5.93KB
更新时间 2026-07-28
语言