Data README

Brook Trout Density Explorer — Subwatershed README

Overview

This document supports the Predicted Brook Trout Density — Eastern United States, Subwatershed (HUC12) Scale Shiny application. The application displays predicted brook trout density across the species’ native eastern U.S. range for 1981–2025 aggregated at the HUC12 subwatershed scale. Estimates are available for Young-of-Year (<100 mm), Adult, and Total brook trout.

The package contains spatial subwatershed boundaries, survey sampling locations, temporal-average predictions, and yearly prediction tables. The Shiny application uses data_huc12.rds for the time-series view and data_huc12_static.rds for the temporal-average view.

Data

We integrate four decades of monitoring data:

  • Presence–Absence: Presence–absence observations from the Eastern Brook Trout Joint Venture (EBTJV) assessment.

  • Abundance: Count data collected via single- and multi-pass backpack electrofishing surveys.

  • Covariates: The presence of rainbow trout, the presence of brown trout, log transformed upstream drainage area, log transformed stream slope, the base flow index, elevation, proportion forest cover, proportion urban development, maximum summer temperature lagged one year, summer precipitation lagged one year, autumn precipitation, and spring precipitation.

Statistical Methodology

The abundance predictions are based on an integrated species distribution model developed by Dr. Justin Van Ee, Dr. Yoichiro Kanno, and Dr. Mevin Hooten. Key features of the modeling framework include:

  • Detection Error Correction: Corrects for detection errors inherent in single- and multi-pass backpack electrofishing.

  • Spatially Varying Coefficients: Estimates heterogeneous covariate effects by watershed (HUC6).

  • Spatial Smoothing: Uses information from neighboring catchments and watersheds to improve predictions in undersampled areas.

  • Uncertainty Quantification: Quantifies prediction uncertainty derived from posterior predictive distributions.

Note: Predictions are intended to display broad spatial and temporal patterns across the landscape rather than exact population estimates for individual stream reaches.

The explorer and underlying dataset are recommended for:

  • Visualizing broadscale spatial differences in brook trout abundance.

  • Inferring temporal trends over the 1981–2025 period.

  • Prioritizing focal populations for conservation and management.

  • Identifying historically undersampled regions to plan future field surveys.

Prediction Variables and Interpretation

Density

The map displays predicted brook trout density using the log10-transformed scale variable mean_log10.

Density values represent the predicted number of brook trout per 100 m of stream.

Uncertainty

  • lwr and upr define the lower and upper bounds of the 90% Credible Interval around predicted density.

Life Stage Categories

  • Young-of-Year (YOY): Brook trout in their first year of life, defined by a length cutoff of <100 mm.

  • Adult: Adult brook trout.

  • Total: Total brook trout density across all age classes.

Data Files

huc12_geom.rds

Spatial polygon layer defining the HUC12 subwatersheds displayed on the map. Each record represents a subwatershed boundary.

Column Definition
huc12 12-digit Hydrologic Unit Code identifying the subwatershed feature.
name Name of the HUC12 subwatershed.
watershed Name of the larger watershed containing the subwatershed.
states U.S. state(s) containing or intersecting the subwatershed.
geometry / Shape Subwatershed polygon geometry stored as an sf object (WGS84 EPSG:4326).

sample_points_geom.rds (Optional)

Spatial locations of brook trout sampling events (mapped as centroids or sample points within a HUC12). The table contains records for survey locations displayed in time-series mode.

Column Definition
huc12 12-digit Hydrologic Unit Code identifying the subwatershed.
year Year in which the sampling event occurred.
sample_type Type of survey represented: Count (shown as circles) or Presence-Absence (shown as triangles).
geometry / Shape Point geometry representing the survey location or subwatershed centroid.

data_huc12_static.rds

Temporal-average prediction table used by the application’s Temporal average (1981–2025) view. Predictions are summarized across the full 1981–2025 period for each HUC12 and life-stage dataset.

Column Definition
huc12 12-digit Hydrologic Unit Code identifying the subwatershed.
dataset Brook trout life-stage category: Young-of-Year, Adult, or Total.
mean_log10 Mean predicted density on the log10 scale. Transformed density is calculated as 10^mean_log10 - 1.
lwr Lower bound of the 90%% credible interval for predicted density.

| | upr | Upper bound of the 90%% credible interval for predicted density.

| | EBTJV_assessment | Eastern Brook Trout Joint Venture (EBTJV) 2024 range-wide assessment status.

| | last_efish_label | Display-ready label describing the most recent count/electrofishing survey. |

data_huc12.rds

Primary long-format prediction table containing yearly estimates for HUC12 subwatersheds and life-stage datasets spanning 1981–2025.

Column Definition
huc12 12-digit Hydrologic Unit Code identifying the subwatershed.
dataset Brook trout life-stage category: Young-of-Year, Adult, or Total.
year Year associated with the prediction (1981–2025).

| | mean_log10 | Mean predicted density on the log10 scale. Transformed density is calculated as 10^mean_log10 - 1. | | lwr | Lower bound of the 90% credible interval for predicted density.

| | upr | Upper bound of the 90% credible interval for predicted density.

| | electrofished | Indicator of whether an electrofishing count survey was conducted in the subwatershed in the selected year (Yes/No).

| | PA_observed | Indicator of whether a presence–absence survey was conducted in the subwatershed in the selected year (Yes/No).

| | EBTJV_assessment | Eastern Brook Trout Joint Venture (EBTJV) 2024 range-wide assessment status.

| | last_efish_label | Display-ready label describing the most recent count/electrofishing survey. |

File Organization & R Usage

brook_trout_huc12_data/
├── README.md
├── huc12_geom.rds
├── sample_points_geom.rds
├── data_huc12.rds
└── data_huc12_static.rds

Read files into R:

library(sf)
library(dplyr)

# Load tabular predictions
data_huc12        <- readRDS("data_huc12.rds")
data_huc12_static <- readRDS("data_huc12_static.rds")

# Load spatial polygons
huc12_geom <- readRDS("huc12_geom.rds") |> st_as_sf()

Data Contributors

Data were contributed by the following state and federal natural resource agencies:

  • Maine Department of Inland Fisheries and Wildlife

  • New Hampshire Fish and Game Department

  • Vermont Fish and Wildlife Department

  • Massachusetts Division of Fisheries and Wildlife

  • Rhode Island Department of Environmental Management

  • Connecticut Department of Energy and Environmental Protection

  • New York State Department of Environmental Conservation

  • Pennsylvania Fish and Boat Commission

  • New Jersey Department of Environmental Protection, Division of Fish and Wildlife

  • Delaware Department of Natural Resources and Environmental Control

  • Maryland Department of Natural Resources

  • West Virginia Division of Natural Resources

  • Virginia Department of Wildlife Resources

  • Tennessee Wildlife Resources Agency

  • North Carolina Wildlife Resources Commission

  • South Carolina Department of Natural Resources

  • Georgia Department of Natural Resources

  • Shenandoah National Park

  • Great Smoky Mountains National Park

Funding

This project was funded by Multistate Conservation Grants (F24AP00078, F25AP00122) from the U.S. Fish and Wildlife Service and jointly administered with the Association of Fish and Wildlife Agencies.