Fig 1.
The domain of the Public Land Survey investigated in this study.
The broad domain includes Minnesota, Wisconsin and the upper two thirds of Michigan state (greyed cells). A 8x8km grid is superimposed over the region to aggregate data, resulting in a total of 7940 cells containing data. The striped band represents the Tension Zone [16], adapted across the region from Andersen [72].
Table 1.
Biomass parameters used for the calculation of biomass in the pre-settlement dataset (rounded for clarity).
Fig 2.
Structural characteristics of PLS era forests.
Total stem density (a) in the Upper Midwest, along with forest type classification (b) based on PLSS data and the stem density thresholds defined by Anderson and Anderson [44] (Prairie: < 0.5 stems ha-1; Savanna: 0.5–47 stems ha-1; Forested: > 47 stems ha-1). Fine lines represent major rivers. To a first order, basal area (c) and biomass (d) show similar patterns to stem density (but see Fig 3).
Fig 3.
The ratio between biomass and stem density as an indicator of forest structure.
Regions with high stem density to biomass ratios (brown) indicate dense stands of smaller trees, while regions with low stem density to biomass ratios (green) indicate larger trees with wider spacings.
Fig 4.
Forest composition (%, using stem density) for the 15 most abundant tree taxa in the PLSS.
The scale is drawn using a square-root transform to emphasize low abundances. Shading of the bar above individual taxon maps indicates plant functional type assignments (dark gray: needleleaved deciduous; light gray: needleleaved evergreen; white: broadleaved deciduous).
Fig 5.
The five dominant forest types in the Upper Midwest as defined by k-medoid clustering.
Forest types (from largest to smallest) include Hemlock-Cedar-Birch-Maple (yellow), Oak-Poplar-Basswood-Maple (light purple), Tamarack-Pine-Spruce-Poplar (light green), Oak Savanna (dark purple) and Pine (orange). These forest types represent meso-scale (64km2) forest associations, rather than local-scale associations.
Fig 6.
Proportional distribution of Plant Functional Types (PFTs) in the upper Midwest from PLSS data.
Broadleaved deciduous tree (BDT), needleleaved deciduous tree (NDT), and needleleaved evergreen tree (NET) distributions are shown as proportions relative to total basal area, total biomass, and composition (Fig 2). The grassland PFT is mapped onto non-tree cells with the assumption that if trees were available surveyors would have sampled them.
Fig 7.
PLSS to FIA comparisons for stem density and basal area.
Scatter plots of the relationship between (a) average stem density and (b) total basal area in the PLSS and FIA datasets. Stem density tends to be higher in the FIA, but total basal area tends to be higher in the PLSS. A 1:1 line has been added to the panels to indicate equality.
Table 2.
Mean cell-wise change in forest zone density and basal area since the PLSS for cells with coverage in both PLSS and FIA eras by forest class.
All forest zones show increases in stem density since the PLSS era (positive values, historical values are included in parentheses). Oak Savanna and the Oak/Poplar/Basswood/Maple are the only zones with increasing basal area since the PLSS, all other zones show declines.
Fig 8.
Distributions of minimum (within dataset) dissimilarities during the PLSS (a) and FIA (b) show spatially structured patterns of dissimilarity, with stronger spatial coherence for the PLS. Lost forests (c) show strong compositional and spatial coherence, and have more taxa with percent composition > 10% than within Novel forests during the FIA era (d). The lost Elm-Oak-Basswood-Ironwood zone lies south of the Tension Zone [16,72], while the other zones lie largely north or east of the Tension Zone. The spatial structure of the novel forests is less well defined.
Table 3.
Spatial distance to novelty—modeled as a binomial—from remnant forests (forests within the first 25th percentile of nearest neighbor distances).
The null model uses permutation (n = 100) where quantiles are resampled without replacement.
Fig 9.
The model relating the probability of novel forests to spatial distance from remnant forest.
Here the 25%ile is used to indicate remnant forest, and the 95%ile is defined as novelty. Binomial regression predicts the probability of novel forest, the red dashed line indicates a response greater than 0.5. The curves represent the relationship between spatial distance and dissimilarity for each of the five major historic forest types (Fig 5) defined here as Oak Savanna (blue), Oak-Poplar-Basswood-Maple (light purple), Tamarack-Pine-Spruce-Poplar (green), Hemlock-Cedar-Birch-Maple (yellow) and Pine (orange). Points are jittered to improve display. Points at 1 are cells whose dissimilarity is greater than the 95th %ile of dissimilarities within the PLSS, here considered novel forest.
Fig 10.
Transects (a) across the region show clear changes in the ecotonal strength.
Transect One shows shifts in broadleaved taxon distributions from the PLSS to FIA (b and c) and in needleleaved distributions (d and e). Transect Two broadleaf (f and g) and needleleaf (h and i) taxa show shifts that again appear to represent regional scale homogenization. Ecotones in the pre-settlement era were stronger in the past than they are in the present. Fitted curves represent smoothed estimates across the transects using Generalized Additive Models using a beta family.