Issues in creating a count history to run N-mixture models

[Juan Camilo Cepeda Duque]

Hi Jurgen, 

Sorry if the following question has been previously solved but I haven't found a relevant way yet while searching in previous inquiries: 

I have collected a series of detections of domestic dogs from a camera trap survey I conducted between september 2022-2023 in the Central Andes of Colombia. We're talking about 23 camera trap stations and a survey effort of 4,657 trap-nights with 1366 raw pictures of domestic dogs. From these pictures, I used digikam to tag every detection event with the number of dogs present in each image. I then used the eventSummaryFunction to obtain counts on the number of dogs per detection event and it provided me with a column in the conventional record database with the number of dogs per detection event. What I haven't found yet, and even searching in the CRAN tutorial available for this package is a way to get my count history to prepare and run an N-mixture model for the dogs in my study area. I also collapsed the database of detection events for 60 minutes and 24 hours, leading the largest number of dogs of all the detections to collapse as the final number of dogs in the collapsed event. For instance, if, in a given day/hour for a given camera trap station, we obtained two detections of dogs and detection 1 had three dogs but detection 2 had four dogs, we lead the number of dogs obtained in detection 2 as the total number of dogs observed for that station on that day. 

Attached are the resulting database ('perro_conteo_crudos_NN1.csv') with the column indicating the number of dogs per detection event, this same database collapsed to 60 minutes ('perros_60min_NN_2023_1.csv') and 24 hours ('perros_24h_NN_2023_1.csv'), and another database of camera operation ('CAM_NUDO_2023_1.csv') I created to construct detection-non-detection histories. 

I would appreciate it if you can help me with this issue.

Thanks in advance,

J. 

--

IUCN SSC Tapir Specialist Group