Plant beta-diversity across biomes captured by imaging spectroscopy

Abstract

Monitoring the rapid and extensive changes in plant species distributions occurring worldwide requires large-scale, continuous and repeated biodiversity assessments. Imaging spectrometers are at the core of novel spaceborne sensor fleets designed for this task, but the degree to which they can capture plant species composition and diversity across ecosystems has yet to be determined. Here we use imaging spectroscopy and vegetation data collected by the National Ecological Observatory Network (NEON) to show that at the landscape level, spectral beta-diversity-calculated directly from spectral images-captures changes in plant species composition across all major biomes in the United States ranging from arctic tundra to tropical forests. At the local level, however, the relationship between spectral alpha- and plant alpha-diversity was positive only at sites with high canopy density and large plant-to-pixel size. Our study demonstrates that changes in plant species composition and diversity can be effectively and reliably assessed with imaging spectroscopy across terrestrial ecosystems at the beta-diversity scale-the spatial scale of spaceborne missions-paving the way for close-to-real-time biodiversity monitoring at the planetary level.

Fig. 1. Location of NEON sites.

The NEON sites used in this study are located (a) across the entire United States and b cover all major biomes except for tropical rainforest. Colors represent different biomes; for site abbreviations and characteristics see Supplementary Table 1.

Fig. 2. Spectral variation among plant communities captures differences in species composition.

a At each site, the average pairwise spectral distance among plots increases with their average pairwise taxonomic distance. Sites are ordered across a latitudinal gradient from north (dark colors) to south (light colors). The overall relationship between pairwise spectral and taxonomic distances across all NEON sites is displayed in golden color and dashed line (n = 13222, r² = 0.18, b = 0.47, t₁₃₂₂₁ = 191.2, P < 0.001). Significance of the relationship between spectral and taxonomic distance was assessed using two-sided t-tests; the number of observations is indicated by the gray tiles in the background. b Covariance between plot-wise ordinations of mean spectra and plant species inventories per site. Colors represent different biomes and stars significance levels, ***P ≤ 0.001, **P ≤ 0.01, *P ≤ 0.05, no star and transparent shading indicates P > 0.05 (not significant). For site abbreviations see Supplementary Table 1, for statistics see Supplementary Tables 2, 3, for results per site see Supplementary Fig. 2.

Fig. 3. Spectral-diversity–plant-diversity relationships at the alpha-scale depend on leaf area index (LAI).

Relationships between spectral alpha-diversity and a plant species richness (LAI 1: n = 157, P = NS; LAI 2: n = 165, r² = 0.08, b = 3.23, t₁₆₃ = 3.96, P < 0.001; LAI 3: n = 149, r² = 0.05, b = 3.10, t₁₄₇ = 3.05, P = 0.003; LAI 4: n = 118, r² = 0.02, b = 2.65, t₁₁₆ = 1.90, P = 0.06), b Shannon index (LAI 1: n = 157, r² = 0.16, b = − 0.17, t₁₅₅ = −5.54, P < 0.001; LAI 2: n = 165, P = NS; LAI 3: n = 149, r² = 0.05, b = 0.21, t₁₄₇ = 2.85, P = 0.005; LAI 4: n = 118, r² = 0.05, b = 0.25, t₁₁₆ = 2.70, P = 0.008), and c phylogenetic diversity (LAI 1: n = 157, P = NS; LAI 2: n = 164, r² = 0.05, b = 0.02, t₁₆₂ = 3.21, P = 0.002; LAI 3: n = 149, r² = 0.04, b = 0.03, t₁₄₇ = 2.60, P = 0.01; LAI 4: n = 118, r² = 0.12, b = 0.06, t₁₁₆ = 4.13, P < 0.001). As phylogenetic diversity measure we used phylogenetic species evenness (PSE). Significance was assessed using two-sided t-tests; colors represent four classes of LAI: LAI 1 = [0.13, 0.634], LAI 2 = (0.634, 1.183], LAI 3 = (1.183, 1.84], LAI 4 = (1.84, 3.80]). For spectral-diversity–plant-diversity relationships at the alpha-scale per site see Supplementary Fig. 3.

Fig. 4. Spectral alpha-diversity predicts plant alpha-diversity.

Correlations between predicted and measured a plant species richness (n = 589, r² = 0.66, b = 1.03, t₅₈₇ = 33.80, P < 0.001), b Shannon index (n = 589, r² = 0.57, b = 1.03, t₅₈₇ = 27.72, P < 0.001), and c phylogenetic diversity (n = 588, r² = 0.40, b = 1.04, t₅₈₆ = 19.87, P < 0.001) per plot calculated from mixed effects models including spectral alpha-diversity, vegetation type (forest, grassland, shrubland), mean leaf area index, latitude, elevation, precipitation and temperature. As phylogenetic diversity measure we used phylogenetic species evenness (PSE). The black line shows the overall model fit, one standard deviation is shaded in gray, the dashed line is the 1:1-line, gray lines are the linear regressions per site, dot colors represent vegetation types. Significance was assessed using two-sided t-tests, no adjustments for multiple comparisons were made; for mixed effects model statistics see Supplementary Tables 4–6.