Evaluating movement-based methods for estimating the frequency and timing of parturition in mule deer

Table 2 Performance of 6 movement-based methods for identifying parturition events and parturition timing for mule deer in Utah during summer of 2023

Method	Accuracy (%)	Precision (7 days)	Sensitivity (%)	Specificity (%)
Peterson	87.7 \(\pm\) 3.2	78.4 \(\pm\) 4.4%	89.8 \(\pm\) 3.1	62.5 \(\pm\) 17.1
BCPA	91.5 \(\pm\) 2.7	34.7 \(\pm\) 4.8%	99.0 \(\pm\) 1.0	0.0 \(\pm 0.0\)
rMCP	91.5 \(\pm\) 2.7	90.3 \(\pm\) 3.1%	94.9 \(\pm\) 2.2	50.0 \(\pm\) 17.7
IBM	58.5 \(\pm\) 4.8	70.7 \(\pm\) 6.0%	59.2 \(\pm\) 5.0	50.0 \(\pm\) 17.7
PBM	90.6 \(\pm\) 2.8	46.8 \(\pm\) 5.1%	98.0 \(\pm\) 1.4	0.0 \(\pm\) 0.0
API	92.5 \(\pm\) 2.6	89.2 \(\pm\) 3.2%	94.9 \(\pm\) 2.2	62.5 \(\pm\) 17.1

Methods tested include the Peterson method, behavioral change point analysis (BCPA), rolling minimum convex polygons (rMCP), individual based method (IBM), population based method (PBM), and the analysis of parturition indicators (API). Accuracy is the percentage of female deer whose parturition status was correctly identified (N = 106), precision is the percentage of parturient deer whose date of parturition was correctly estimated within 7 days of the true date of parturition (N = 98), sensitivity is the percentage of parturient deer who were correctly identified as parturient (N = 98), and specificity is the percentage of barren deer who were correctly identified as non-parturient (N = 8)

ISSN: 2051-3933