Tuesday, July 28, 2009
Stock and Statistics
It has always been fascinating to mix two seemingly immiscible concepts with a firm objective in mind. Statistical ogling of fish biological dynamics happens to be the best suited case under this category. One of the immediate moot points in this line is the stock based inference arrived through quasi-linear or non-linear modeling. Statistics being inductive in nature is bound to blow up the results to some universe and "stock" fits the bill perfectly. But there are non-statistical issues on the very definition of an unit stock. The wikipedia throws up a dozen such definitions and points to the most acceptable one which unfortunately happens to be the latest. (http://en.wikipedia.org/wiki/Fish_stock) this has the potential danger of the query- "what happened to previous inferences". In the sampling concept a sample and population are uniquely, indisputably defined and the methodology sticks on that perfect framework. But it seems to be not the case in Stock assessment. Further stocks by definition vary with the part of the globe in which the study is being conducted. Another issue is the collation of various stock based inferences- are the additive? or some interaction is to be accounted for? Are the pelagic and demersal assessments even footed in terms of setting?
Subscribe to:
Post Comments (Atom)
It is no surprise that ( being a hardcore statistician) your views on stock assessment are rather queer, to put it mildly. Being a elite member of stock assessment group of a reputed and well known Institute you are sure aware of the concept of Unit Stock and its characteristics and which in my opinion has a clear definition. Correct me if I am wrong. Fish stocks may vary but what constitutes a stock does not vary. Your last lines are not clear to me. Do you mean adding MSYs of individual stocks in a region and giving a TOTAL MSY for the region? Well! if it is so then they are not additive. I am lost on your pelagic and demersals.
ReplyDeleteN.B. : In the sampling concept a sample and population are uniquely, indisputably defined and the methodology sticks on that perfect framework. But it seems to be not the case in Stock assessment --- Give me some specific examples. Please.
It was an innocuous query which arose out of inquisitiveness arising while narrating the process of assessing(estimating parameters of) a population (stock). It was felt that the concept of stock which has been put in various forms and words needs to be categorically put when it comes to inferring about the same using mathematical/ statistical procedures. Begg et al (1999) in their definition cited in the wikipedia link talk of "semi-discrete group". Probably this has some clue.
ReplyDeleteHere also is the FAO glossary link for stock: http://www.fao.org/fi/glossary/default.asp
ReplyDeleteHowever, the question of additivity or probably rather interactivity of stocks in a region is a really intersting one that is drawing mmore and more attention. Though so much stock assessment is based on single species concepts, or their extrapolation to multiple species fisheries, some arguments for and against lumping stocks (within and across species) can be given. However, at the 2008 5th World Fisheries Congress in Yokohama, one of the new papers on harvest rules pointed out that no country will ever be able to have sufficient information on even the most important species to enable management on the basis of separate genetic stocks. So catch limits or advice will need to be precautionary and contain guidance on how to spread the catch and effort spatially if possible, in case total catch that is safe for certain genetic stocks of a species is concentrated in a limited area and thereby removes a particular genetic stock. This is thought to have already happened with some of the genetic stocks of cod in the Atlantic, and has likely removed some stocks with specific behaviours and climate preferences, thus reducing the resilience of the species under climate change.
However, more importantly for highly mutli-sepcies tropical fisheries, perhaps 'ecosystem level' fishing or management is a concept worth pursuing?
I have the least professional knowledge in stock assessment.....I need a clarification from anyone of you....Is there an empirically vaildated stock assessment model specific to pelagics? Is it true that all the dominant models in stock assessment have taken demersal spp for model building and validation studies?
ReplyDeleteAlso, whether EBFM or MSY our management tool box is having hardly four or five options. And ultimately the decision is political ...( I don't find anything wrong in it,Science is for the people and may be to the fish also through the people)There is an urgent need to liberate the tropical fisheries management epistemology from the tyranny of temperate epistemological moorings....
Ramchandran
ramchandrancnair@gmail.com
I appreciate the pelagic concerns of C. Points well made and taken. To my knowledge there doesnot seem to exist any "empirically validated stock assessment model" especially for small pelagics in tropical waters. Most of the models employed even in demersal stock assessment depend on so many assumptions, which in most cases are untenable. But, results from those stock assessment excercises can be taken as first approximations to the status of the stocks conditional to the validity of the assumptions. In my view, one must be able to project the pessimistic and optimistic projections of the stocks and also take into consideration the recent trends in exploitation and the landings(catches).
ReplyDeleteThat, MSY may not be desirable management reference point has been demonstrated in many cases. There are other reference points worth exploring. In tropical waters where natural mortality rate is relatively high, the classical models may fail to arrive at an estimate of MSY and even if they do indicate it will be at a high rate of fishing. Yes, as C has indicated There is an urgent need to liberate the tropical fisheries management epistemology ( or philosophy) from the tyranny of temperate epistemological moorings....
For those who are serious about stock assessment I recommend the following books. They are top class and give a clear insight to stock assessment models and approaches and fisheries management under uncertainty. If you have a weak heart for maths( both the books have plenty of it), skip maths read the explanations.
ReplyDelete1).Quantitative Fisheries Stock Assessment: Choice, Dynamics and Uncertainty
by R. Hilborn and C.J. Walters
Kluwer Academic Publishers
2).The Ecological Detective:
Confronting Models with Data
Ray Hilborn & Marc Mangel
Princeton University Press ( for Hardcover)
Springer ( for Paperback)
It is a monograph.
An extract from
ReplyDeleteSchnute, J.T., L.J. Richards, and N. Olsen. 1998. Statistics, software,
and fish stock assessment. Pages 171–184 in Funk, F., T.J. Quinn II, J.
Heifetz, J.N. Ianelli, J.E. Powers, J.F. Schweigert, P.J. Sullivan, and C.-I.
Zhang (Editors). 1998. Fishery Stock Assessment Models. Alaska Sea Grant
College Program Report No. AK-SG-98-01. University of Alaska Fairbanks.
Proceedings of the International Symposium on Fishery Stock Assessment
Models for the 21st Century. October 8–11, 1997. Anchorage, Alaska.
Lowell Wakefield Fisheries Symposium.
-------------------------------------------
Although modern statistical and software tools have become quite sophisticated, no elaborate model should preclude a thoughtful examination of the facts. Fishers often accuse scientists of placing excessive faith in mathematical models and ignoring practical experience. We regard models as tools for investigating biological scenarios consistent with the facts, analogous with a detective constructing the crime scene from available
clues. A credible model scenario can also be supported by an intuitive examination of the raw data . The final product must go beyond the mathematical paradigm to an explanation that policy makers can understand and communicate to a skeptical public.
-----------------------------------------------
I request Dr.JJ and his colleagues to read the full paper available at
http://otter-rsch.ca/SchnuteRichardsOlsen1998.pdf
Thanks for the excellent references suggested by Dr Srinath and Dr Meryl.
ReplyDeletean immediate observation on the paper by worm et al , thanks to Meryl again, is the apparent lack of stock data representation from the Indian ocean region (fig1) on page 579. I would be grateful if anyone could tell me why it is so....
Ramchandran
My immediate response to C's immediate observation is that we do not have a time series data on stock sizes even of the major exploited species. It does not reflect badly on the capablities of research workers in this field simply beacuse that the data we could afford to collect on the these species cannot generate the kind of output C has mentioned. Surely, the data generated do indeed tell us about the historical and probable future trend of the fishery and its output. Another important point is that most of the resources that are exploited form the seas around India are considered as annual crops, mostly supported by 0-year are less than 1-year class.
ReplyDeleteMoreover, the models employed in Worm's paper to estimate Bmsy or Bmmsy and other management reference points are questionable for obvious reasons. The goodness of fit of these models needs to be critically examined. The validity of the single species approches in a multi-species system needs a critical revisit. ( I think there are n number of papers on this). I have strong reservations on EBF Models which are data hungry and whose requirements and assumptions could rarely be met. Another point to note the inputs used in these models are inappropriately aggregated over species and estimates of Z or M are based on certaing strong assumption whose validity in an interacting complex system is debatable.
Thanks for the excellent clarification provided by Dr Srinath. I think it would be nice if you could send a remark on the paper by Worms et al to Science. It is high time people of your calibre make a crtique in these lines. It would dfenitley help to invite more international research attention or caution while dealing with tropical fisheries
ReplyDeleteOf late I am reading the New Testament and I have come across interesting passages related to "discards". Paranthesis mine:
ReplyDelete"Again, the Kingdom of heaven is like a dragnet that was cast into the sea and gathered some of every kind.
Which, when it was full, they drew to shore; they sat down and gathered the good into vessels and threw the bad away(unwanted ones!)" -- Mathew 13-47