Movement analysis for Mara/Grumeti data
-
Data Merge Takes cleaned tracking datasets and merges into a single dataframe NOTE: Grumeti data is at 30-min intervals!! #TODO: Add the grumeti metadata in the GR cleaner code #TODO: Add region names to collar metadata - should be possible with DAS
-
Raster Extraction Raster extraction for the full dataset. Limited to ecosystem-wide layers.
-
MovData Filter Filter out individuals based on overlap with spatial layer extents and time (must overlap at least one crop season)
-
Tactic Clustering
GMM/Tactic Clustering Calculates ag use stats (mean, 90-day max, etc.) Applies GMM models to the full data set Cross validation of model Option to filter and apply to only a single data set
GMM/GMM_cluster_results Get cluster cutpoints and classify individual-year tactics Figures and tables summarizing aggregate and individual-year tactic cluster results
GMM/Ag_Regression_Models Regression models of ag use in relation to movement and sex/age characteristics
GMM/Tactic_Change_Models Regression models on tactic change
- HMM Fit HMM, evaluate, simulation, and activity budgets
HMM_prepData_pop_GME Preps data and calculates log speed. Filters out individuals with high % missing fixes
HMM_population_fit_GME Fit HMM candidate models
HMM model assess plots & Simulation Assess the model outputs
HMM_ActivityBudgets_GME Apply viterbi algorithm to estimate latent states Calculate activity time budgets and plot (Fig S11) Calculate activity density bugets and plot (Fig 4) Conduct overlap tests and apply regression models