\section{Figure documentation}\label{app:figures}
All figures are generated by \texttt{figures/make\_figures.py} (matplotlib, vector PDF) with the validated palette of \texttt{figures/style.py}; the table lists, for each figure, its caption and the result artefacts it reads. Colours follow a fixed categorical order; one axis per panel; direct labels for at most four series; the 2022-11-30 reference date is drawn as a dashed line and never as a series.

\begin{longtable}{p{0.22\linewidth}p{0.46\linewidth}p{0.26\linewidth}}
\caption{Figure inventory.}\label{tab:figure-inventory}\\
\toprule \textbf{File} & \textbf{Caption} & \textbf{Sources} \\ \midrule \endfirsthead
\toprule \textbf{File} & \textbf{Caption} & \textbf{Sources} \\ \midrule \endhead
\texttt{fig\_prisma\_flow.pdf} & PRISMA-style flow of record identification, exclusion and the final mapping corpus. & prisma\_counts.json \\
\texttt{fig\_volume\_yearly.pdf} & Eligible works per publication year by source membership (2026 incomplete). & volume\_yearly.csv \\
\texttt{fig\_volume\_monthly\_raw.pdf} & Monthly first appearances; January spikes reflect year-only publication dates and motivate the month-precision filter. & volume\_monthly.csv \\
\texttt{fig\_types\_openness.pdf} & Document-type composition and metadata completeness by year. & volume\_yearly.csv \\
\texttt{fig\_collaboration.pdf} & Team size, international co-authorship and industry participation by year. & team\_size\_yearly.csv; industry\_share\_yearly.csv \\
\texttt{fig\_citation\_inequality.pdf} & Fixed-window citation levels and inequality by publication year (years with complete 36-month windows). & citation\_windows\_yearly.csv \\
\texttt{fig\_field\_inflows.pdf} & Share of references pointing outside the corpus, by cited OpenAlex field and citing year. & field\_flows\_yearly.csv \\
\texttt{fig\_cited\_age.pdf} & Distribution of the age of cited works for selected citing years. & cited\_age\_yearly.csv \\
\texttt{fig\_venue\_concentration.pdf} & Number of distinct venues and venue concentration (HHI, top-source share) by year. & venue\_hhi\_yearly.csv \\
\texttt{fig\_country\_shares.pdf} & Country shares of output with fractional counting, top eight countries. & countries\_yearly.csv \\
\texttt{fig\_its\_total.pdf} & Monthly output with the interrupted-time-series fit; level and slope changes with HAC and placebo p-values. & fit\_total.csv; its\_results.json \\
\texttt{fig\_placebo\_total.pdf} & Placebo distribution of ITS t-statistics over every candidate event month; the true event is highlighted. & placebo\_total.csv \\
\texttt{fig\_category\_growth.pdf} & Yearly primary-category output indexed to 2018, fastest- and slowest-growing categories. & category\_counts\_yearly.csv \\
\texttt{fig\_area\_shares.pdf} & Share of eligible works by top-level area (primary category) per year. & category\_counts\_yearly.csv \\
\texttt{fig\_breaks\_by\_category.pdf} & ITS slope change at 2022-12 by category with level change annotated; bars in colour pass the placebo test. & breaks.csv \\
\texttt{fig\_area\_flow\_matrix.pdf} & Citation-flow matrices between areas (row shares) for the baseline and post periods. & area\_flows\_yearly.csv \\
\texttt{fig\_cross\_area.pdf} & Cross-area reference shares by citing area and modularity of the category citation network by year. & cross\_area\_share\_yearly.csv; modularity\_yearly.csv \\
\texttt{fig\_emerging\_metrics.pdf} & Out-of-time detection performance of the emerging-topic models by horizon. & metrics\_h12.json; metrics\_h24.json \\
\texttt{fig\_known\_set.pdf} & Monthly keyword series for the twelve preregistered concepts with the detector's first alarm. & known\_set\_monthly.csv \\
\bottomrule
\end{longtable}