Supplement 1. Delphi/Pascal scripts for simulating and parameterizing individual-based movement models.

File List
Hypercube_21_8_3x10e6_10e5.txt

IN_real_bas.txt

IN_real_galera.txt

IN_real_galera98.txt

IN_real_monteria.txt

IN_real_palas.txt

IN_real_Plantilla.txt

IN_real_quin.txt

IN_real_sierrecica.txt

MOD1_Definitions.txt

MOD1_Demographics.txt

MOD1_Dispersal.txt

MOD1_Main.txt

MOD1_Movement.txt

MOD1_Parameterization.txt

MOD1_Utils.txt

MOD2_Definitions.txt

MOD2_Demographics.txt

MOD2_Dispersal.txt

MOD2_Main.txt

MOD2_Movement.txt

MOD2_Parametrization.txt

MOD2_Utils.txt

usos4bas10m.txt

usos4gal10m.txt

usos4gal98_10m.txt

usos4pal10m.txt

usos4quin10m.txt

usos4sierr10m.txt

Description
Indivual-based models were developed in Delphi programming language. Programs in Delphi are divided in units, each one represented by a .txt file (in Delphi, with extension '.pas'). Our programs have six units with name starting with 'MOD_'. These units use external information:

- Radiotraking record of radiotracked individuals (observed patterns). One file per location. Name: 'IN_….txt'

- Parameterizations that will be tested. One file pr program. Name: 'Hypercube_….txt'

- Habitat maps of the study areas (matrix of values). Name: 'usos4….txt'

We present two programs:

- Model1 (MOD1_) is the main simulation model that yields for each parameterization,

for each observed individual i and summary statistic j two indices m_ij and v_ij

that quantify the deviance between the summary statistics of the observed and simulated data considering observation error and process stochasticity

(see also Fig. D1 and Appendix D for details).

The dimensions (cols × rows) of the matrix result of this program = (n × 8 × 2) × (NPARM) where n × 8 × 2 means number of individuals × 8 summary statistics × 2 indices (mij, vij), and NPARM means the number of parameterizations tested.

- Model2 (MOD2_) is a modified version of Model1 that assesses the inter-individual variability and the stochasticity due to the sampling protocol. Model2 describes, for all pairs of individuals, noise in the observed data with respect to the mean and standard deviation of the summary statistic j (m_j_kl, v_j_kl, see Appendices for details). Starting from these values, threshold values tmj, tvj are calculated.