Spatial proximity moderates genotype uncertainty in genetic tagging studies.

Area of Science:

• Ecology
• Conservation Biology
• Genetics

Background:

• Global animal population declines necessitate improved demographic estimation methods.
• Current noninvasive genetic methods for population density estimation are inefficient due to unusable low-quality samples and identification errors.
• Accurate population density and trajectory estimates are crucial for effective wildlife conservation.

Purpose of the Study:

• To introduce the genotype spatial partial identity model (gSPIM) for more reliable and efficient demographic parameter estimation.
• To reduce genotype uncertainty and increase the precision of population density estimates by integrating spatial and genetic information.
• To demonstrate the model's effectiveness in improving accuracy and efficiency in wildlife population monitoring.

Main Methods:

• Developed the genotype spatial partial identity model (gSPIM) by integrating genetic classification and spatial population models.
• Applied the gSPIM to noninvasive genetic samples from a fisher (Pekania pennanti) population study.
• Utilized simulation studies to assess the model's accuracy, precision, and parameter identifiability.

Main Results:

• The gSPIM increased density estimate precision by 25% and corrected three individual identity assignment errors in the fisher study.
• Simulation studies showed a 63% increase in accuracy and a 42% increase in precision for density estimates.
• Model parameters were identifiable with a single replicated assignment per sample, with accuracy and precision being relatively insensitive to replication for high-quality samples.

Conclusions:

• The gSPIM significantly enhances the accuracy and precision of population density estimates from noninvasive genetic data.
• The model allows for the utilization of previously discarded lower-quality samples, improving data yield and efficiency.
• Genotyping protocols can be optimized by reallocating resources from high-quality to low-quality samples when using the gSPIM.