Abstract
Wild pigs (Sus scrofa) are a highly destructive invasive species across
the globe, making them subject to intensive management including lethal
control. Robust estimates of population abundance are needed to assess
and optimize the effectiveness of control efforts. Genetic capture
mark-recapture (GCMR) offers considerable promise for monitoring
invasive wild pigs. However, obtaining robust estimates from GCMR can be
difficult due to low quality DNA, particularly in moist, hot
environments that promote fast DNA degradation. To examine if GCMR is
feasible for estimating wild pig abundance, we collected pig fecal
samples in three sites (bottomland hardwoods, mixed forest, and upland
pine) at the Savannah River Site, South Carolina, USA. Amplification
success across nine microsatellite loci varied across habitats with
bottomland hardwoods having the lowest success (18%) compared to the
mixed forest (56%) and upland pine (65%). Resultant abundance and
density estimates were relatively similar between field-based methods
and GCMR in the bottomland hardwoods and upland habitats, but estimates
for the bottomland hardwoods had larger confidence intervals. Tests with
additional extraction methods in the bottomland hardwoods found low
amplification, with a combination of Nucleospin soil kits and Zymo
clean-up kits performing the best. While our study found reasonable
estimates of density across three habitats, environmental conditions
have a powerful influence on amplification success and the corresponding
number of recaptures in wild pig GCMR, particularly in the bottomland
hardwoods where flooding was frequent during sampling. Successful GCMR
studies should, therefore, consider both sampling intensity and
laboratory costs when designing studies because dryer habitats have
higher amplification success but lower pig densities.