This is an interesting discussion and something we've been grappling with at Conservation Evidence. We have begun a project to build an AI-assisted evidence synthesis pipeline using LLMs primarily to make the process of evidence synthesis more efficient as a workflow (from finding and classifying relevant scientific studies testing conservation actions, to tagging key information to speed up the writing of evidence summaries). The ultimate goal is to build a living evidence database that is able to keep up the the rapidly growing scientific literature.

Specifically regarding Chat-gpt (at least 3.5) we have found it's very poor at providing evidence-based answers to questions on conservation. The most worrying thing is that if decision-makers are using it out-of-the-box, they may feel they're getting an authoritative answer and this may exaggerate the issue of overconfidence. It often makes up sources or at least suggests it got info from a study that isn't relevant. Of course this may change with future iterations, but currently I fear it's being used without proper safeguards or knowledge of its limitations, specifically for decision support. We are trying to build a more credible NL interface, fine-tuned on the CE database that will have a built in verification model that checks the evidence sources provided, and would tailor answers based on a user's location. The challenge is reducing hallucinations as much as possible and whether ethically this is still acceptable. Ultimately, my feeling is if we don't try to build something more credible, with better safeguards, people will end up using naïve LLMs that are worse and will lead to bad decision-making.

Hello! 

I am a Conservation Programs Manager for Assiniboine Park Conservancy (Winnipeg, MB, Canada) focused mostly on Arctic and Northern species. My main expertise is in the world of biologging for marine mammals, but I also work with other techniques like camera trapping, photogrammetry, drones, and more! I also support other conservation projects related to critically endangered grassland species, facilitating conservation planning, and supporting Indigenous and community-led research. 

I've been on this site for a number of years, but realized I never introduced myself to this amazing community!

Courtney

Hello Wildlabs community!

I’m excited to be here and introduce myself. I work for the Wildlife Conservation Society (WCS) in Argentina and have recently joined the Wildlabs team. I’m a biologist with a strong interest in bioacoustics and conservation. I've studied cetaceans in Argentina and Antarctica, using passive acoustic tools and visual methods to monitor these populations. My PhD focused on the acoustic behavior of Commerson’s dolphins and the impact of vessel noise in Patagonia, Argentina.

My current role involves conducting a global horizon scan on bioacoustics, aiming to assess the current state of bioacoustics and identify gaps, challenges, and priorities to advance the field, with a long-term vision.

I’m also part of the GIAS Research Group at the Universidad Nacional de Tres de Febrero, which aims to develop software and hardware to facilitate bioacoustic research projects in Argentina. Additionally, I teach at this university, offering a seminar on underwater acoustics to sound engineering students to spark their interest in bioacoustics and wildlife conservation. I’m thrilled to be part of this community and look forward to contributing and learning with all of you!

Hi everyone, I just got my first hydromoth and wanted to test it for aquatic soundscape with interest in Tomistoma, Otter, boat traffic and maybe fishes too!  But before that I maybe test it on zoos.

What are your advices, tips, or suggestion for first-time user? thank you!

You won't get any audio if you don't allow enough time for the hydromoth/audiomoth to write. So when you do a continuous recording you need to experiment a little. I'm sure there is a formula to calculate this, but I haven't figure that out. I typically do 5 min recordings with 10 seconds of write/break time. I think this system is expecting you to subsample, so keep that in mind instead of a continuous recording. 

I do 8kHz sampling and get about 7 days of data and then the voltage gets too low and you start getting SD card write errors and missing files. 

In terms of analysis, I've had trouble understanding the directionality of the hydromoth and incorporating this into my studies. I always set up two at the same site to check the variability in my call detections and include this into my error analysis. 

Sounds interesting. 
We all used acoustic membranes.
Just like @BrettMargoSupplies suggested.
A friend of mine had really bad results with animal borne
mics where the pcb was potted and the mic was just covered by a thin layer of epoxy where it hit the surface.
That was not working good!

Greetings from thhe woods,
Robin

Congratulations, very exciting! Keep us updated!

This is so cool @Mauricio_Akmentins - congrats and look forward to seeing your project evolve!

Congratulations! My first hydromoth was just arrived yesterday and so excited! Looking forward for the update from your project!!!

Hello @CourtneyShuert 

If I am not mistaken, Wildlife Computers was developing or has developed a remote release package for such an application as has a much smaller company, Desert Star Systems. I do not know where either company stands in terms of functionality, but I think for both cases the animal still has to carry some kind of an attachment plate that will eventually be shed off. In theory (purely theory) and assuming a product exists, you/they could consider adapting the material of the plate to something that is 'sustainable' and not plastic or similar.

(edit: I noticed Kyler linked the WC product above)