Late-successional and old-growth forest across the Northeast: definition, mapping, and the temporal (in)stability of satellite embeddings

Aggregated results, analysis code, and figures supporting a Northeast late-successional and old-growth (LSOG) forest methods paper. The study defines LSOG through a multi-axis structural label built on U.S. Forest Inventory and Analysis (FIA) data, produces design-based four-state estimates (Maine, New Hampshire, Vermont, New York) with a New Brunswick provincial-inventory extension, maps the probability of LSOG from AlphaEarth satellite embeddings, and assesses the temporal stability of those embedding-based predictions. Headline findings. (1) Design-based FIA estimates place any-LSOG at roughly 15% of forest in Maine, 28% in New Hampshire, 28% in Vermont, and 33% in New York (2019-2023 panel). (2) A single-time LSOG classifier trained on the multi-axis structural label discriminates well against reserve-anchored references (AUC ~0.84). (3) Canopy-height resolution and AlphaEarth-plus-canopy-height fusion tests show only marginal gains over AlphaEarth alone. (4) The temporal-stability analysis is the substantive addition in this version: AlphaEarth embedding vectors are highly stable year to year (median cosine similarity ~0.98), a classifier refit independently each year shows no temporal decay (pooled per-year AUC ~0.64 to 0.65), yet year-to-year embedding change does not track disturbance at the plot scale (change-detection AUC ~0.45, at chance). The practical implication is that plot-level embedding differencing is unreliable for change detection, so LSOG trends should rest on design-based FIA rather than on differencing satellite embeddings between years. Associated outputs. These aggregated results and code underpin two CRSF technical reports, a four-state regional report and a Maine-focused report (Weiskittel, Woodall and D'Amato 2026), and two peer-reviewed manuscripts in review: a methods paper (Ecological Applications) and a Comment on Hagan et al. 2026 (Ecosphere). Restricted data. FIA true plot coordinates, New Brunswick MagPlot per-plot records, and JDI ownership layers are NOT included and remain restricted; only aggregated summaries (H3 hexagon and state-level tables), analysis code, and figures are provided. The lead file 00_LSOG_NE_graphical_abstract.png is a graphical abstract summarizing the study.