Camera trap survey-specific data - where to put it?

Support
,
1

What Wildbook are you working in? ACW

I’m looking for help determining the best fields to use in the system for a particular dataset type. The org we’re working with gets all of their data from camera trap studies. They have the following categories they’d like to be able to track with particular sets of images:

  • survey area: specific survey project with a large, defined geographical area
  • study area: a subset of the survey area
  • survey year: some survey &/or study areas are re-used over multiple years so being able to identify which year the data was collected in that area is important

Having a way to recognize, at the encounter record level, which survey & or study area each encounter belongs to, along with the survey year, is what we’re trying to find a solution for. Being able to do this in the bulk import would be optimal, because that’s how their data will be loaded into the system.

In ACW, we haven’t yet tried to use any of the “survey”-related fields such as:

  • Occurrence.fieldStudySite
  • Occurrence.fieldSurveyCode
  • SurveyTrack.vesselID (for obvious reasons)
    but these all seem to be at the Occurrence level not at the Encounter level.

Other options I’ve considered as part of the solution are:

  • Occurrence ID: include some of the survey info in the OccID string but it’s likely not practical as a survey is much broader than a sighting
  • Projects: it’s logical to consider a survey a project in ACW but an encounter being part of a specific project is not displayed at the encounter level, if I’m not mistaken. Also, I’m not sure the projects functionality is robust enough yet, particularly around location info? But maybe it is, or is heading in that direction?
  • Location ID + verbatim locality: to capture survey & study areas. Not sure this is practical since I don’t believe the survey & study areas necessarily follow defined geographical locations and are generally at a much lower level than country or region. Also, some study/survey areas are only used once, such as when a local community grants 1-time access for a cameras to be placed on their land.

We know of many camera trap surveys underway with numerous ACW users and future users, so this is something we want to solve for in the most optimal way possible that everyone doing these surveys can use going forward.

I’m assuming that this has likely been solved for in other Wildbooks so I’m hoping to learn from others’ experience in figuring this out. Thoughts and recommendations would be very much appreciated.

thanks!
Maureen

2

Hey there,
I just dug through all my notes on how people are tracking similar things and here are my best recommendations and responses to what you have:

  1. Survey fields are limited and I don’t recommend using them. We want to get them built out well but the full use case isn’t defined so they may get some changes that would make them radically different from your intent.
  2. Project is visible at the encounter level. It’s under “Metadata” since technically it’s not ecological data. This would be my top recommendation out of the choices you’ve listed.
  3. Occurrence ID: This would likely lead to a lot of duplication across your OccID strings because, as you pointed out, a survey is more than one sighting (hopefully). However, if you could find a naming convention that worked well, it would be doable.
  4. Location ID + verbatim locality: This is my second highest recommendation. You can maintain location IDs as the broader regions, but if something is used frequently, such as a specific park, this could be very useful. Vebatim locality would be great for the study area; this is the only one I have seen used by others for something like this (most camera trap entries focus on the GPS location, so this is actually a fairly new use case).
3

Perfect! This works for me. Thanks tons @tanyastere!

4

hi @tanyastere, I’ve been mulling this and one thing stands out as a possible constraint re: using projects functionality for these camera trap surveys - that’s the ID’d individuals are by project ID. What happens when an individual is seen across multiple surveys? Basically, I’m trying to work out what is the right level of survey info to put at the Wildbook project level to avoid having multiple ID’d individuals across multiple projects?

Could the project ID part of the individual ID be considered a temporary ID? I.e. after seen across multiple surveys, can I change the ID name to something that doesn’t reference the project ID that the individual got id’d in?

Also, how does the system reconcile a pre-existing, non-project associated, ID dataset that gets uploaded as part of an existing ID kit for that organization, with a project ID? I’m guessing the merge functionality but I’m not clear on how that would manifest. But I REALLY want to use project functionality for cam trap surveys - it just feels like exactly what’s needed. Maybe if I had a better understanding of the use case(s) that Project functionality was created to support???

Your thoughts on this are VERY much appreciated!!!

Below is the specific problem I’m working to solve for and likely representative of additional cam trap surveys in Africa with other orgs (I already know of another, cross-border, multi-species, multi-organization example that headed our way).

CCF explains their camera trap approach as follows:
We have multiple study areas in Namibia, and in the not so distant future, in Somalia and possibly other countries also.

Within each study area, surveys take place either once only, or are replicated. The number of replicates per study area differs - e.g., from 2 to >10. The time interval between replicates differs also (e.g., depending on study area and survey, can be seasonal, annual, or every few years).

The survey types differ also: e.g., cheetah-specific surveys may involve camera traps placed at mark trees. Whereas mammalian biodiversity surveys may involve camera traps placed along roads, trails, as well as mark trees. We also have regular monitoring in specific areas for waterholes etc.

Ex1.
CCF: Organization acronym
GWLElan: Acronym for Greater Waterberg Landscape study area + First 4 letters of farm name (Elandsvreugde) where cam trap photo was taken
WHole: Survey details, in this case long-term monitoring of waterholes
KuduDam: Camera station name

Ex2.
CCF: Organization acronym
GoSwee: Acronym for Gobabis study area + First 4 letters of farm name (Sweetwater) where cam trap photo was taken
Svy2021-2022: Survey details, in this case survey carried out in 2021-2021
C16Trail: Camera station name (in cell 16 on trail)

Thanks
Maureen