1
Deepak George Pazhayamadoma, Emer Rogana, Ciaran Kellyb and Edward Codlingc
aSchool of Biological, Earth and Environmental Sciences (BEES), University College Cork, Ireland; bFisheries Science Services, Marine Institute, Ireland; cDepartment of Mathematical Sciences, University of Essex, United Kingdom
Can we manage a fishery if no previous data are available?
Application of quality control charts in management of data limited fisheries
Historical data
Yes
Qualitative risk assessments
Quantitative stock assessments
No
Self Starting Cumulative Sum�SS-CUSUM
YES
No historical data at 0th year
2
SS-CUSUM
Self starting CUSUM (Hawkins, 1998)
Running mean�(Calibrated using real time data)
Three parameters
1. Allowance (k) �2. Control limit (h) �3. Winsorizing constant (w)
Parameters
Evaluation of SS-CUSUM using a stochastic simulation test
Indicator observations corresponding to out-of-control situations are omitted while calibrating the running mean
Performance measures used
3
Results (ROC curves)
Conclusion
All stock indicators in the study were useful in detecting fishing impact and hence SS-CUSUM can be potentially used for monitoring data poor fisheries
Reference:
Hawkins, D.,Olwell, D., 1998. Cumulative sum charts and charting for quality improvement: Springer Verlag, pp:162-168.
Best
Good
Worst