Estimating Steelhead Redd Abundance and Variance in the

Estimating Steelhead Redd Abundance and Variance in the Wenatchee River Basin Andrew Murdoch WDFW Chad Herring WDFW Kevin See QCI Chris Jordan NOAA

Agenda n Background n Objectives n Methods n Preliminary results n Plans for 2012

Background n UCR Steelhead listed in 1999 n Need estimate of spawner abundance to estimate productivity n UCR steelhead, like most others, have a huge variation in migration patterns “Overshooting” natal streams n Hatchery strays n Differential overwintering mortality n n Tributary versus Columbia

Goals and Objectives n Goal n Estimate the total number of redds with an estimate of variance n Objective Develop models to estimate redd observer efficiency based on environmental, biological, or habitat variables n Estimate number of redds using AUC and redd life n Estimate variance based on total uncertainty in the models n

Survey Methodology n Conduct weekly surveys of index areas Geo-reference all redds n Collect environmental data each survey n Conduct habitat survey of each index area (one time) n n Conduct single survey of non-index areas at or near peak spawning of associated index area n Collect environmental and habitat data

Index and Non-index reaches n Index reaches are not random Includes all major spawning areas n Number of index reaches depends on spawn timing and redd life n n Non-index reaches Everywhere else n Same spawn timing and redd life as index reach n

Redd Life Summary n Steelhead spawning timing progresses from low to upper elevations, redd life is positively correlated with elevation. n Redd life is only as good as survey frequency May require some intensive surveys in the first few years n Some surveys may need to be performed after spawning is complete n

2010 Steelhead Redd Life Reach Mean N SD CV P 1 I 1 W 9 W 10 N 1 N 3 16. 2 19. 5 18. 2 18. 0 15. 1 20. 4 44 47 38 50 10 13 8. 7 6. 7 9 9. 1 8. 1 5 53. 5 34. 4 49. 5 50. 6 53. 6 24. 5 Range Min 3 8 3 3 5 14 Max 41 32 45 45 28 33 % of all redds 70% 42% 36% 36% 15%

Observer Efficiency Results by Reach Census reach N Redd statistics Mean CV Range Redds % SD Omission False ID % SD P 1 9 16 38 3 - 24 48. 8 18. 3 51. 2 18. 3 17. 2 17. 6 N 3 10 34 45 17 - 54 61. 2 16. 2 38. 8 16. 2 14. 0 11. 3 I 1 10 24 29 12 - 36 49. 8 14. 3 50. 2 14. 3 22. 5 9. 0 W 9 8 21 77 9 - 52 30. 9 22. 3 69. 1 22. 3 40. 6 11. 7 W 10 11 34 40 18 - 61 41. 2 13. 1 58. 8 13. 1 20. 7 13. 2

Overall Best Model Reach Experience 2. Visibility 3. Redd density 4. Channel Complexity Index 1.

Statistical Analysis – Redd Abundance Estimate of redds by reach for both index and non-index areas n AUC n n Use normal approximation of redd counts through time Apply some of the same parameters to non-index reaches Get estimate of total number of redd-days Divide by redd life n Divide by observer efficiency n

Index Reach (Icicle River)

Non-Index Reach (W 3)

Statistical Analysis – Variance n Sources of variance Redd life n Observer efficiency n Parameter uncertainty in estimating the normal curve n n All sources can be included using the delta method

Conclusions n Observer efficiency can be estimated reasonably well n Methodology can be used with a reasonable amount of effort n Redd life is important, but somewhat subjective n Training, training

Next Steps n Collect more observer efficiency data n n Maximize contrast to the greatest extent possible Evaluate importance of survey effort (hours) into model Evaluate alternate metrics for channel complexity (thalweg variation) Model validation n Conduct weekly surveys on all census reaches Estimate redd abundance and variance for the Wenatchee Basin in 2012 n Conduct same study in the Methow using two surveyors instead of one (2012 -2014) n

Acknowledgements Funding provided by BPA and Chelan County PUD n All the WDFW spawning ground surveyors n Chris Jordan conceived the project n Jody White and Chris Beasley at QCI for valuable insight into the project n Kevin See for graciously accepting the project with no warning n