The Agreement on International Humane Trapping Standards (AIHTS) which applies to wildlife management, vertebrate pest control, and trapping for fur, skin, or meat for 19 listed species requires that a trapping method render at least 80% of a minimum of 12 target animals irreversibly insensible within a species-specific time limit. However, the Agreement also allows for the use of other scientifically proven methods as a substitute for testing on live animals. For the past five years, we have been developing computer models and simulation systems to determine whether killing traps meet humane trapping standards. The models were designed to classify the time-to-loss-of-sensibility of furbearing species based on mechanical characteristics of traps and strike location(s). Models were based on data collected from trap testing on marten (Martes americana), fisher (Martes pennanti), and raccoon (Procyon lotor). Models were tested against 15 years of live trap testing data from the Fur Institute of Canada. The models proved to be a valid alternative to trap testing on live animals due to their high levels of safe prediction accuracy (88%, 86% , and 92% for marten, fisher, and raccoon, respectively). If applied to trap testing, these models would dramatically reduce the cost and the need for trap testing on live animals.

We conducted field trials during 2000 to compare the effectiveness of 0.4% strychnine commercial ready-to-use (RTU) hull-less oat and canary seed (CS) bait to control Richardson’s ground squirrels (Spermophilus richardsonii) and during 2001 to compare the effectiveness of RTU and freshly prepared from concentrate (FFC) hull-less oat bait. Each study was conducted at 30 locations in ungrazed pastures and unharvested forage crops in southern and central Alberta from mid-June to mid-August. Effectiveness was measured using pre and post treatment visual, dead and in 2000 active burrow counts. In both trials visual counts increased with daily maximum temperature (P < 0.05), decreased with rainfall intensity (P < 0.001) and were not affected by wind speed or start time (P > 0.05). The 2000 field trials indicated that adjusted visual counts of ground squirrels were significantly lower than controls in CS than RTU baited plots (77.6% and 59.6% reduction, respectively, P = 0.002) and the mean number of dead ground squirrels was higher (6.43 and 2.13, respectively, P < 0.001). Additionally, there were significantly (P < 0.001) fewer re-opened holes by ground squirrels in the RTU (41.6%) and CS (71.7%) treated plots compared to control plots. All 3 measures indicated greater effectiveness using canary seed than hull-less oat bait. The 2001 field trials indicated that adjusted visual counts of ground squirrels were significantly lower than controls in FFC than RTU baited plots (92.7% and 65.6% reductions, respectively, P < 0.0001) and the adjusted mean number of dead ground squirrels was higher (4.28 and 1.21, respectively, P < 0.0001). Both measures indicated greater effectiveness using freshly prepared from concentrate than manufactured ready to use hull-less oat strychnine baits. In conclusion, we advocate further research into freshness as bait freshness may be the single most important factor affecting effectiveness