2014 CA Salmon forecasts

http://www.pcouncil.org/wp-content/uploads/Preseason_Report_I_2014_FINAL.pdf

I just skimmed it.  It doesn't appear the greatest.  Sac Fall ... 634K vs 834K last year.  Klamath Fall ... 299K vs. 727K last year.   

-Allen

I expect a full report from you on all 151 pages on my desk by tomorrow morning.

BMB

Who but a biologist could grasp this fully?  From just skimming the stats I agree it doesn't appear as good as last year but not a disaster either.  Ben is there gonna be a pop quiz on this?

It says we get to fish for salmon this year
""In 2014, invoking de minimis fishing rates under Amendment 16 will be unnecessary because SRFC
potential spawner abundance is projected to be greater than 162,667 hatchery and natural area adults.""


Interesting note about the method used to calculate the Sacramento Index (SI) copy & pasted below:
""Prior to 2013, the SI had been frequently over forecast, prompting an investigation into alternative models
and data that could be used to improve forecast accuracy. The analysis evaluated the forecast
performance of a wide range of models and data sources and compared performance of these alternative
approaches to the conventional SI forecast model. The results suggested that there was scope for modest
improvements in SI forecast performance, leading the STT to recommend a new SI forecasting approach
for use in the future.""
""The new approach continues to use jack escapement estimates to predict the SI, but uses a wider range of
data and accounts for autocorrelation in model errors. The accounting for autocorrelated errors is the
most substantial change to the SI forecast method used prior to 2014. In practice, this means that if in the
previous year the modeled SI value was larger than the SI postseason estimate for that year, the SI
forecast is adjusted downward to account for that error. Conversely, if the modeled SI value in the
previous year was less than the postseason estimate of the SI for that year, the SI forecast would be
adjusted upward to compensate for that error.""
""For comparison, use of the previous SI forecasting approach would result in a 2014 SI forecast of
476,342.""
""In 2013, the preseason forecast of 834,208 was 97 percent of its postseason value of 862,525.""

Yeah I saw the note about the over forecast.  When you take a look at the recent numbers you start to wonder if the model is also over-forecasting for the Klamath fishery as well.

I think it's cool they adjusted the equation--and it's providing more accurate forecasts, so far.
SI predictions in recent years:
2008: 54K   
2009: 122K   
2010: 245K   
2011: 729K   
2012: 819K   
2013: 834K   
2014: 634K

Actual numbers:
2008: 69K
2009: 41K
2010: 150K
2011: 206K
2012: 624K
2013: 862K

Yeah I saw the note about the over forecast.  When you take a look at the recent numbers you start to wonder if the model is also over-forecasting for the Klamath fishery as well.

Was it really overforecasting?

""In 2013, the preseason forecast of 834,208 was 97 percent of its postseason value of 862,525.""

Although it doesn't give me much confidence when the old model would have forecast 476K and the new one is forecasting 634K.

-Allen

Yeah I saw the note about the over forecast.  When you take a look at the recent numbers you start to wonder if the model is also over-forecasting for the Klamath fishery as well.

Was it really overforecasting?

Just going by the report for the recent years, from 2007-2013, the preseason forecast has been higher than the postseason estimate by a ratio of 1.20 - 1.88.  Those seem to show that the forecast has been at least slightly optimistic over those years.  Of course that means little due to the sample size being small, but I do think its interesting.  Not sure what the drivers were (if any) and I wouldn't even wager to guess. 

But when you factor in that they so much as admitted the Sac forecast was off for the same years, it's worth a question at least.  But that's the scientists' job, not some hack on the internet like me.

Ben,  where those numbers from the report?  Why are the numbers Pat posted different?

-Allen

I think Pat's numbers were for the Sacramento river. They use SI as the index for the sac. My numbers were for the Big K and pulled directly from the report. Sorry for any confusion.

Sent from my Galaxy Nexus using Tapatalk

seriously Ben, got to get the numbers right here.  This is important stuff and its how the PFMC decides to jerk us around when the state isn't and the feds let the fish have a little water.

JK, 

My list was just the Sac. Ben's list is the Klamath. 

Article in Sac Bee this AM that explains it to all us statistically challenged folks.  Basically, not as good as last year, but plenty of fish for a full season.

Sent from my Nexus 7 using Tapatalk

Yes! I'm so happy! I have a job this summer! This morning I've already tied up 50 hootchie/ flasher rigs and am about to start on 100 spoon spreads.

Salmon results 2013:

Raydon v. Salmon              0 for 5.   

I'm hoping for a better ratio for 2014.