The scoring methodology translates merged species distribution models and extinction risk weights into cell-level sensitivity metrics, which are then rescaled and aggregated to management zones for decision support.
For each grid cell, sensitivity is computed as the extinction risk weighted sum of species presence within each species category (Equation 7.1):
\[ \text{score}_{c,g} = \sum_{s=1}^{S_g} \frac{\text{er\_score}_s \times v_{s,c}}{100} \tag{7.1}\]
where:
Division by 100 converts the product from a 0–10,000 range back to a 0–100 scale.
In plain terms, for each cell in the ocean the MST adds up the sensitivity contributions of all species found there. If a cell has many species that are both likely to be present and at high risk of extinction, it gets a higher sensitivity score. This helps identify places where rare or threatened species are concentrated.
Cell-level scores are stored as named metrics in the DuckDB database:
| Metric Key | Description |
|---|---|
extrisk_bird |
extinction risk score for birds |
extrisk_coral |
extinction risk score for corals |
extrisk_fish |
extinction risk score for fishes |
extrisk_invertebrate |
extinction risk score for invertebrates |
extrisk_mammal |
extinction risk score for marine mammals |
extrisk_reptile |
extinction risk score for reptiles/turtles |
extrisk_other |
extinction risk score for other marine organisms |
primprod |
primary productivity (VGPM, metric tons C km-2 yr-1) |
Raw cell scores vary naturally across ecoregions due to differences in species richness, oceanographic conditions, and biogeographic patterns. To enable meaningful cross-region comparison, each metric is rescaled to a [0–100] range within each BOEM ecoregion:
\[ \text{score}'_{c} = \frac{\text{score}_{c} - \text{score}_{min}}{\text{score}_{max} - \text{score}_{min}} \times 100 \tag{7.2}\]
where \(\text{score}_{min}\) and \(\text{score}_{max}\) are the minimum and maximum cell values for that metric within the ecoregion. This ensures that a score of 100 represents the most sensitive cell within each ecoregion, not globally.
Net primary productivity (NPP) from the VGPM satellite model is included as a separate metric (primprod) that does not undergo extinction risk weighting. Instead, the raw NPP values (metric tons C km-2 yr-1) are directly rescaled within each ecoregion using the same [0–100] normalization.
Rescaled cell scores are aggregated to three spatial levels using area-weighted averaging:
For each zone, the aggregated score is:
\[ \text{score}_{zone} = \frac{\sum_{c \in zone} \text{score}'_c \times A_c}{\sum_{c \in zone} A_c} \tag{7.3}\]
where \(A_c\) is the area of cell \(c\) (accounting for latitude-dependent cell size in the 0.05° grid).
The flower plot provides an intuitive summary of sensitivity scores by species category, inspired by the Ocean Health Index (Halpern et al. 2012) visualization approach:
Figure 7.1 shows flower plots for all 20 BOEM Program Areas, generated from the v3 scoring run. Each flower summarizes the sensitivity profile of one program area with petals for Bird, Coral, Fish, Invertebrate, Mammal, Other, Turtle, and Primary Production. Alaska program areas (e.g., HOP, COK, GOA) show dominant Fish and Mammal petals, while Gulf areas (GAB, CEC, MAT) show stronger Invertebrate and Coral contributions.
The flower plot lets decision-makers quickly spot which ecological elements drive sensitivity in an area, aiding better planning and impact assessment.
The flower plot is implemented as an interactive ggiraph visualization using polar coordinates in R (ggplot2 + coord_polar()), with tooltips showing detailed score breakdowns on hover. The plot_flower() function in calc_scores.qmd generates static versions for reports, while the mapgl web application implements an interactive JavaScript version for responsive rendering (see Figure A.2).
The treemap provides a detailed, hierarchical view of species contributions to sensitivity scores (Figure 7.2):
Zooming into a specific category reveals individual species contributions:
The primary visualization tools are two Shiny web applications: