NEWS.md

In ct: Camera Trap Data Management and Multi-Method Ecological Analysis

ct 0.5.0

New functions

• ct_fit_encounter_response(): tests whether a response species is detected more or less often at a camera in the hours after a trigger species passes. Comes with print() and summary() methods.
• ct_plot_calendar(): calendar heatmap of daily detections.
• ct_fit_distribution(): fits a Poisson, negative binomial or binomial distribution to counts and reports AIC and BIC.

New datasets

• lama: detections and deployments from 23 cameras in the Lama Classified Forest, Benin, with distances and angles.
• penessoulou: image records from the Penessoulou Classified Forest, Benin. Replaces the two CSV files previously in inst/.

Improvements

• ct_fit_ds(): much faster bootstrap for binned distance data (fast_bootstrap = TRUE, the default).
• ct_correct_datetime(): new time_zone argument, and optional end times to correct clock drift linearly.
• ct_overlap_matrix(): bootstrap confidence intervals (n_boot, conf, ci_method, cores).
• ct_describe_df(): new by_group argument.
• ct_plot_overlap(): new species_names and legend_title arguments; rug ticks are now coloured by species (rug_color to override).
• ct_plot_camtrap_activity(): new legend_title argument.
• ct_check_location(): columns can be given unquoted.

Installation

• sf and terra move to Suggests. ct_spatial_coverage(), ct_survey_design(), ct_solartime() and ct_check_location() ask for them when called.
• Minimum Rcpp version lowered from 1.0.14 to 1.0.8.

ct 0.4.0

2026-06-20

• ct_temporal_shift() now also returns Displacement (in hour): the signed shift of the activity window along the day, measured at its midpoint (positive = later, negative = earlier). This captures a pure time shift, which Shift size (a change in window duration) reports as ~0.
• ct_temporal_shift() gains period_names and legend_title arguments to set the legend labels (e.g. c("Dry", "Rainy")) and legend title directly, instead of the fixed "First period"/"Second period"/"Period".
• Fixed a major performance bug in ct_fit_ds() bootstrapping. Distance::bootdht() re-resolves a model's symbolic call arguments with parent.frame(n = 3), which misfires because ct_fit_ds() calls it from one stack frame deeper: arguments such as cutpoints failed to resolve, so each bootstrap replicate silently dropped the distance binning and fell back to the far slower exact-distance likelihood (observed ~19x slowdown, e.g. ~25 min vs ~1.3 min for one replicate). The model's stored call is now frozen to literal values before bootstrapping, so the bootstrap refits on the intended binned data.
• ct_fit_ds() gains a seed argument.
• ct_fit_ds() now shows a progress bar with an ETA during bootstrapping when the progress package is installed and n_cores == 1. When n_cores > 1, it reports up front that live progress is unavailable (a Distance limitation), so a long parallel run is not mistaken for a freeze.
• ct_fit_rest() Fit the Random Encounter and Staying Time (REST / RAD-REST) model
• ct_fit_tte(), ct_fit_ste(), and ct_fit_ise() for Time To Event (TTE), Space To EVent (STE), and Instantaneous Sampling Estimator (ISE) respectively for density/abundance estimation.

ct 0.3.0

2025-08-09

• Added Distance Sampling functions:
• ct_fit_ds() for fitting detection functions and estimating density/abundance.
• ct_availability() for temporal availability corrections.

• ct_QAIC(), ct_chi2_select(), and ct_select_model() for automated two-stage model selection.

• Added Camera Trap Data Package (Camtrap DP) integration:

• ct_dp_read() to load Camtrap DP datasets from local files or URLs.
• ct_dp_table() to access specific tables (observations, deployments, media, events, taxa).
• ct_dp_example() to load example dataset.
• ct_dp_version() to retrieve dataset standard version.
• ct_dp_filter() to subset tables using dplyr-style filtering.

ct 0.2.0

2025-07-29

Improved ct_stack_df() - C++ implementation for stacking a list of data frames.

2025-07-10

Added new functions to support trap rate and REM-based density estimation workflows: ct_traprate_estimate() estimates trap rates from detection data; ct_fit_activity() models diel activity patterns; ct_fit_speedmodel() fits animal movement speed models; ct_fit_detmodel() estimates detection probability functions; ct_fit_rem() applies the Random Encounter Model (REM) to estimate animal density; ct_get_effort() calculates sampling effort metrics such as camera-days; and ct_traprate_data() prepares detection and effort data for further analysis.

2025-06-26

• ct_correct_datetime() to correct datetime stamps in camera trap datasets using a deployment-specific correction table. Supports multiple datetime formats, offset directions.

2025-06-25

• ct_plot_camtrap_activity() function to visualize camera trap deployment activity with optional gap indicators.
• ct_summarise_camtrap_activity() function to compute summary statistics for camera trap deployment activity, including active durations, gaps, and activity rates, etc.

2025-06-24

• Improved handling of non-numeric variables in ct_describe_df().
• Added support for detecting sampling breaks using ct_find_break().
• Added function to compute confidence intervals (ct_ci() and ct_lognorm_ci())
• Fixed NSE-related warnings

First Release Highlights

• Initial release of maimer
• Provides tidyverse-friendly functions for data cleaning, transformation, and visualization.
• Includes support for alpha & beta diversity, species activity overlap, and temporal analysis.
• Integrates with ggplot2 for customizable visualizations.
• Features an interactive Shiny app for image metadata handling

ct documentation built on Oct. 22, 2026, 1:07 a.m.