WILD HORSE IDENTIFICATION APP

The Problem

Humane population management for wild horses to prevent overpopulation and habitat destruction in areas threatened by development or resource competition. The federal government (Bureau of Land Management - BLM) spends $80 million a year on a program the National Academy of Sciences in 2013 called "expensive and unproductive for the BLM and the public it serves." The NAS said that the current practice of rounding up and removing wild horses from the range is making the problem worse by artificially increasing the reproductive rates of the horses left on the range. Our program, which has already delivered over 1,800 fertility control treatments to more than 1,000 wild mares in Nevada, seeks to expand the program to a national footprint by improving its ability to identify individual horses for more efficient delivery of immunocontraception. This will ultimately stabilize herd populations and reduce their impacts on the ecosystem.

Our Proposal

The program utilizes the PZP immuncontraceptive vaccine delivered remotely by dart rifle to female horses (mares). The vaccine creates an immune response that prevents fertilization but maintains natural behaviors. In the first year, each mare requires a primer and a booster shot; annual boosters are delivered thereafter. Current identification of individual horses and darting history is tracked through a data base that relies on human spotters who evaluate horse color, markings and social affiliation to identify mares in need of treatment. A modern app that would allow identification of mares via technology akin to facial recognition would revolutionize the program and reduce the time it takes to identify individual mares in need of treatment. This would ensure that a sufficient number of mares are treated to reduce population growth rates. The Wildbook platform has already been used to identify individual whales, zebras, elephants, and many other species but not for wild horses. We want to adapt this technology or something similar to our program and believe this is necessary foundational technology to scale up the fertility control program nationwide.

We Assume that...

Ability to generate a large enough data set in a format that can be analyzed

Existence of a unique feature or combination of features to enable individual horse identification.

Technical capacity, as demonstrated by Wildbook and similar apps, exists to develop a horse identification app

Constraints to Overcome

Preparing a large enough data set to utilize artificial intelligence to identify a unique feature that serves as a "fingerprint" to identify individuals within a wild horse herd.

Current Work

- Finalize user stories for darters and documenters - Raise $100,000 to fund development on wildbook or similar platform. - Select appropriate technology platform for the app. - Continue to contribute data and images to existing wild horse information management database. - Engage development team to build app

Current Needs

- Total estimated project budget is $100,000. We seek to raise half of that from external sources. - App developer experienced with wildbook or similar platform - Online storage, approximately 1TB max based on current storage profile. - 2 Internet enabled satellite phones to cover areas not served by cellular coverage today.