Estimate Fleet-Wide Activity from Sample Data
estimate_fleet_activity.Rd
Estimates total fleet activity by scaling sample-based trip statistics using boat registry data. Calculates sampling rates and confidence levels for the estimates.
Arguments
- monthly_stats
Data frame from
calculate_monthly_trip_stats()
containing:district: District name
date_month: Month as date
sample_total_trips: Total trips from tracked boats
sample_boats_tracked: Number of tracked boats
avg_trip_duration_hrs: Average trip duration
avg_trips_per_boat_per_month: Average trips per boat
- boat_registry
Data frame with columns:
district: District name (must match monthly_stats)
total_boats: Total number of boats registered in district
Value
A data frame combining monthly statistics with fleet estimates:
district: District name
date_month: Month as date
sample_total_trips: Total trips from tracked boats
sample_boats_tracked: Number of tracked boats
avg_trip_duration_hrs: Average trip duration
avg_trips_per_boat_per_month: Average trips per boat
estimated_total_trips: Estimated trips for entire fleet
sampling_rate: Proportion of fleet tracked (0-1)
estimate_confidence: Confidence level ("High", "Medium", "Low")
Details
This function combines monthly trip statistics from GPS-tracked boats with boat registry data to estimate fleet-wide fishing activity. It calculates:
Estimated total trips if all boats were tracked
Sampling rate (tracked boats / total boats)
Confidence levels based on sampling coverage
Confidence levels are assigned as:
High: ≥30% of fleet tracked
Medium: 10-29% of fleet tracked
Low: <10% of fleet tracked
See also
calculate_monthly_trip_stats()
for generating input monthly statisticscalculate_district_totals()
for combining with catch/revenue data
Examples
if (FALSE) { # \dontrun{
# First calculate monthly stats
monthly_stats <- calculate_monthly_trip_stats(trips_data)
# Then estimate fleet activity
fleet_estimates <- estimate_fleet_activity(monthly_stats, boat_registry)
# Check confidence levels
fleet_estimates |>
dplyr::count(estimate_confidence)
} # }