[Access this dataset on Dryad]https://doi.org/10.5061/dryad.n2z34tn4x

Please see methods section of manuscript for details on data collection.

Description of the data and file structure

We have submitted data on our localities and quarries (Table S1), our taxonomic scheme and morphotypes (Table S2), and our raw data (Tables S3-S6), as well as our R script (R Code.R), and supplementary figures as referenced in the manuscript text (Figures S1-S3).

Supplement - Tables S1-S6.xlsx: contains supplementary Tables S1 through S6 as described in text

Table S1: quarry information

• UWBM Locality Number: locality number assigned by University of Washington Burke Museum; unique to each quarry and collection year
• Locality Name: locality name as assigned by collector(s)
• Year: year in which collection was made
• Land Ownership: land owner of the property containing the locality, as of the year of collection
• Permit number: permit number assigned by appropriate agency (where applicable) under which fossil specimens were collected

Table S2: morphotype guide (information, diagnostic features, and character scores for each morphotype; all morphotypes and sub-morphotypes are listed in order to document full range of morphological diversity)

• Morphotype Number: number assigned by authors of this study to distinguish this operational taxonomic unit or morphotype
• Sub-Morphotype: sub-morphotype assigned by authors of this study (where applicable); used by these authors to highlight instances where a morphotype includes a broad range of morphological variability
• Diagnostic features: plain text description of distinguishing features used to identify this morphotype
• HQI/MQI: quality index, describes how well preserved the fossil is (see text for more information)
• Morphotype name: currently valid taxonomic name (where applicable)
• In other designations: synonymous morphotype numbers or species names used by other authors in previous studies
• Major plant group: major plant taxonomic group to which this morphotype belongs. Abbreviations: CON = conifer, CYC = cycadophyte, DIC = dicot (non-monocotyledonous angiosperm), GIN = ginkgophyte, MON = monocotyledonous angiosperm, PTE = non-seed vascular plant (pteridophyte), indet. = indeterminate
• Organ type: plant organ preserved in this morphotype
• CIC: compendium index category (see The Compendium Index of North American Mesozoic and Cenozoic Type Fossil Plants, published by Peabody Museum of Natural History; Shusheng Hu, 2012)
• Primary vein framework…teeth per cm: character traits for each morphotype, defined and scored following the framework of Ellis et al. (2009) [Ellis, B., D. C. Daly, L. J. Hickey, K. R. Johnson, J. D. Mitchell, P. Wilf, and S. L. Wing. 2009. Manual of leaf architecture. Cornell University Press, Ithaca.]

Table S3: dataset 1 abundance data (including all taxa): raw count data of abundance of each morphotype at each locality

Table S4: dataset 1 species information (Data from Table S2 transformed for use in analyses in this paper. Only a few data fields differ from Table S4 (defined below). Only the 123 named morphotypes are scored in this table. Blank fields are left intentionally empty.)

• Organ_1 and Organ_2: plant organ preserved in this morphotype (as in Table S2) broken into more or less granular definitions
• Singletons: scored for each morphotype as “Singleton” (only present at one locality) or “Not Singleton” (present at multiple localities)
• Color: design graphic parameter assigned for use in R script

Table S5: locality information

• Site: locality name
• siteno: numeral assigned to the site to place in temporal order
• StratigraphicHeight: distance from the K/Pg boundary in local section (negative being below, positive being above)
• Lithology: localities assigned to channel or overbank depositional settings based on geologic observation (see text)
• KPg: localities scored as K (Cretaceous) or Pg (Paleogene)
• Color and Shape: design graphic parameters assigned for use in R script
• Bin: localities were placed into bins dividing the local stratigraphic sequence into 15-m bins across the K/Pg boundary
• bottom, top, and median:  lowest, highest, and median (middle) stratigraphic height, incorporating calculated error as described in text (used to draw error bars in R script)
• Age: localities scored as HC (Cretaceous), TM1 (earliest Paleocene), or TM2 (early-middle Paleocene) as described in text

Table S6: taxon ID for individual specimens in this study, used for random permutation tests to calculate error on diversity metrics

• ID Number: specimen number (assigned by University of Washington Burke Museum)
• Locality Name: locality name
• Morphotype: assigned morphotype identification
• Singleton_Subset, Organ_Subset, and Taxon_Subset: descriptors and categories for the assigned morphotype of that specimen
• SiteNo: site number (refers to SiteNo assigned to each locality in Table S5)

Table S7: datasets used in each analysis (documents which set or subset of species abundance data was used in each analysis and the figures, tables, or text documenting results; also described in Methods and in figure captions)

Supplement - R Code.R: R code for all analyses performed in manuscript, as outlined in Methods section

Annotations are provided throughout the script through 1) set up environment, 2) import data and transform, 3) diversity metrics, 4) NMDS, 5) DISPER, 6) ANOSIM, 7) Mantel Test, 8) CCA, 9) DCA, 10) iNEXT Richness Analyses, and 11) subsampling for error calculation.

Supplement - Figure S1.png: Stratigraphic sections at each of the 11 localities in this study. Position of each section relative to the K/Pg boundary at left. Stratigraphic sections (1–11) cover the 2–7 m of stratigraphy around the quarry. Each section is measured in meters (at left of the section column) and displays the lithology (see legend for lithologic categories), contact type between lithologic units, names of marker beds (where known), and indication of fossils found at different stratigraphic levels (plant, shell, or vertebrate material). Width of the lithologic units indicates the relative grain size.

Supplement - Figure S2.png: Taxonomic turnover and diversity through our study interval at each of the 11 floras, as in Figure 7 except in Fig. S2 analyses were conducted using all morphotypes. Last Appearance (A) is the percentage of taxa that end their stratigraphic range in that flora. First appearance (B) is the percentage of taxa that begin their stratigraphic range in that flora. Because the proportion of appearances (A–B) are influenced by the start and end of our sequence, we have excluded the oldest and youngest floras where appropriate. Range-through richness (C) is the number of taxa with range-through occurrences in that flora. Rarefied richness (D) is subsampling of abundance data to the smallest sample size (N = 173). On the left, the formations are indicated along with the magnetostratigraphy (based on Sprain et al. 2015, 2018), temporal intervals used in this study (HC, TM1, and TM2), and names of floral assemblages (as in Figure 1). Values are plotted at the stratigraphic height of the corresponding flora, marker shape represents sedimentary facies, vertical lines on markers represent stratigraphic uncertainty, and horizontal lines represent 95% confidence intervals based on bootstrapping of data to N = 173 specimens.

Supplement - Figure S3.png: Taxonomic turnover and diversity through our study interval at each of the 11 floras, as in Figure 7 except in Fig. S3 analyses were conducted using only leaf morphotypes. Last Appearance (A) is the percentage of taxa that end their stratigraphic range in that flora. First appearance (B) is the percentage of taxa that begin their stratigraphic range in that flora. Because the proportion of appearances (A–B) are influenced by the start and end of our sequence, we have excluded the oldest and youngest floras where appropriate. Range-through richness (C) is the number of taxa with range-through occurrences in that flora. Rarefied richness (D) is subsampling of abundance data to the smallest sample size (N = 143). On the left, the formations are indicated along with the magnetostratigraphy (based on Sprain et al. 2015, 2018), temporal intervals used in this study (HC, TM1, and TM2), and names of floral assemblages (as in Figure 1). Values are plotted at the stratigraphic height of the corresponding flora, marker shape represents sedimentary facies, vertical lines on markers represent stratigraphic uncertainty, and horizontal lines represent 95% confidence intervals based on bootstrapping of data to N = 143 specimens.

Sharing/Access information

All specimens are reposited at the University of Washington Burke Museum of Natural History and Culture (UWBM).

Code/Software

Script was written in R version 4.0.2 (R Core Team, 2020; http://www.r-project.org) using the community ecology package vegan version 2.5–7 (Okansen et al. 2019), the fossil diversity dynamics package divDyn (Kocsis et al. 2019), and the species diversity package iNEXT (Hsieh et al. 2016).