EFFECT OF COYOTE REMOVAL ON SHEEP DEPREDATION IN NORTHERN CALIFORNIA

We used 13 years of historical data to investigate effects of coyote ( Canis latrans) removal on depredation of domestic sheep. The 2,168-ha study area maintained >1,000 breeding ewes that produced lambs yearly. Records from 1981 through 1994, which included numbers of sheep, numb ers of sheep known killed by coyotes, known numbers of coyotes removed , and annual numbers of trapper hours were summarized and analyzed on a yearly, seasonal, and monthly basis. We used regression analysis and found that annual, seasonal, or monthly depredation losses were not c orrelated with number of coyotes removed. Both annual number of lambs killed and number of coyotes removed were positively correlated with n umber of trapper hours. We used a cross-correlation analysis to detect any relation between coyote removal and subsequent depredation losses at all monthly intervals from 0 to 24 months. We found a trend of low negative correlation between depredation losses and number of coyotes removed for lags of 2-12 months, suggesting some reduction of sheep k illing due to control efforts. Low correlations within years may be du e to inconsistent removal of depredating coyotes while removing primar ily young, nondepredating coyotes. Lack of correlation between years m ay have occurred because past control efforts have not had a lasting r eduction on coyote density due to immigration, the compensatory nature of control efforts on coyote mortality, reproductive compensation in the resident coyote population, or all 3 factors.