Hydraulic sampling protocol to estimate natural embryo mortality of pink salmon

Productivity of naturally spawning salmon populations is sometimes estimate d by sampling incubating eggs from the spawning grounds to assess survival by pumping water or an air-water mixture into the stream gravel and collect ing eggs displaced by the resulting agitation. General knowledge regarding embryo sensitivity to physical shock indicates that this procedure may kill eggs if conducted before embryo development has advanced beyond blastopore closure, which is up to day 12 postfertilization. In 1998, we assessed the impact of sample timing on egg mortality estimates of pink salmon Oncorhyn chus gorbuscha in 10 spawning streams in Prince William Sound, Alaska, by h ydraulically sampling the same streams in September and again in October. T he mean egg mortality was significantly higher in the early sampling period (63.6%) than in the later sampling period (22.8%). Results from examining stage of advancement at the time of embryo death indicate that sampling wit h hydraulic techniques will elevate mortality among embryos that have advan ced in development up to the early eyed stage, corresponding to day 20 in e mbryo development or approximately 50% longer than previously thought. To a void induced embryo fatality, the sampling protocol should delay assessment of spawning ground survival until sufficient time has past to assure that the most recently spawned eggs have progressed well, into the eye pigmentat ion stage.