WORKING PAPER SERIES FISHERIES STOCK ASSESSMENT TITLE XI1 Collaborative Research Support Program Fisheries Stock Assessrnent CRSP Management Office Internationai Programs, College of Agriculture The University of Maryland, College Park, Maryland 20742
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
WORKING PAPER SERIES
FISHERIES STOCK ASSESSMENT TITLE XI1
Collaborative Research Support Program
Fisheries Stock Assessrnent CRSP Management Office Internationai Programs, College of Agriculture
The University of Maryland, College Park, Maryland 20742
In cooperation with the United States Agency for International Development (Grant No.DAN-4146.G-SS-5071-,)0) the Fisheries Stock Assessment CRSP involves the following participating institutions:
The Univeis-t, of Maryland-Center for Environmental and Estuarine Studies The University of Rihode Island- International Center for Marine Resource DevelopmentThe University of Washington--Center foi Quantitative Sciences The University of Costa Rica-Centro de Investigaci6n en Ciencias del Mar y LimnologiaThe University of the Philippines.-Marine Science Institute (Diliman)--College of Fish
eries (Visayas)
In collaboration with The University of Delaware: The University of Maryland-Collegeof Business and Management; The University of Miami; and The International Center forLiving Aquatic Resources Management (ICLARM).
Working Paper Series
Working Paper No. 9 "Hydrcacoustics and Ground Truth"
by Richard E. Thorne
School of Fisheries University of Washington
May, 1987
Fisheries Stock Assessment Title XII
Collaborative Research Support Program
The Fisheries Stock Assessment CPSP (sponsored in part by USAID GraiL No.DAN-4146-G-SS-5071-00) is intended to support collaborative research betweenU.S. qnd developing countries' universities and institutions on fisheries stock assessment and managenent stratrjies.
This Working Pdoer has been produced by the collaborative efforts of TheUniversity of Washington and The University of Costa Rica-Centro deInvestigacion en Ciencias del Mar y Limnologia (CIMAR) in association with TheUniversity of Delaware. Additional copies are available from the CRSP Management Office and from:
Dr. Vincent Gallucci, Director Center for Quantitative Science HR-20
School of Fisheries University of Washington
3737 15th Avenue, N.E., Room 304 Seattle, Washington 98195
HYDROACOUSTICS AND GROUND TRUTH
Richard E. Thorne
School of Fisheries
University of Washington
Seattle, WA 98195
Abstract
One of the most difficult problems in hydroacousticapplications is the acquisition of ground truth data. A major incentive for the use of hydroacoustics is the deficiencies of the same alternate techniques that areavailable for such ground truth data. However, such data areoften required for biological information including speciescomposition. In addition, since hydroacoustic techniques are not universally accepted, ground truth information issometimes needed to establish confidence in the results. Theauthor has been involved in over 2000 hydroacoustic surveysof various species in a variety of locations. Many different approaches were used to obtain ground truth data. Examination of these results provides some interestinginsights into the characteristics of both the hydroacousticand the ground truth techniques and suggests some principlesfor ground truth efforts in conjunction with hydroacousticsurveys. A major conclusion is that the appropriate role of ground truth efforts is not verification, but is to aid interpretation of the hydroacoustic data.
2
HYDROACOUSTICS AND GROUND TRUTH
Richard E. Thorne
Introduction
I had an occasion in 1980 to compile a list of the
number and type of hydroacoustic surveys in which I had
participated during the previous decade. The total came to
1025 (Thorne 1983). In the subsequent seven years, the total
has probably doubled. These represent a considerable variety
of habitat and species. Virtually all have had some type of
ground truth effort in conjunction with the surveys. The
type and effectiveness of the ground truth efforts were also
highly variable. The purpose of this paper is to summarize
these results and to develop some general principles for
ground truth programs. It should be emphasized that the
results represent my particular experiences. Even 2000
surveys are a small fraction of the potential interactions
among habitats, species and ground truth techniques. Other
investigators may have differing experiences. Nevertheless,
I think the results can provide vaiuable insight into ground
truth techniques.
Ground Truth and the Laws of Physics
Different investigators have differing definitions of
the concept of ground truth. One commonly held concept is
that of verification. Under this definition, cround truth is
an alternate technique that can be used to determine if the
3
hydroacoustic techniques are providing the "right" answer.
This was the concept underlying my initial research in this
field nearly 20 years ago (Thorne 19*!0, L971). One
difficulty with this concept is that there are very few
circumstances where alternate techniques can provide an
adequate standard. The glaring deficiencies of these
techniques are part of the motivation to use hydroacoustics
in the first place. In addition, most fisheries acousticians
have come to realize in the past two decades that the
question is not whether hydroacoustics are "right" or
"wronq", but the appropriate question is, "how can the
information contained in the physics of the acoustic data be
most effecti.rely processed and interpreted?" This concept
starts with the realization that the informaLion contained in
the acoustic data already limits the scope of possibilities
for interpretation. The data already contain considerable
information on the location and nature of targets. The role
of ground truth should be to facilitate interpretation of
this information. This is the concept of ground truth which
forms the basis of this paper.
Scope of Paper
The major techniques that have used in association with
my hydroacoustic surveys have been various types of trawls,
or layerc-d distribution, near-boundary orientation, size of
targets and diel changes in distribution and behavior.
This capaui Iity leads to the fourth pi.nciple: that
ground truth t.echniques are most effective when they are
directed by the acoustic observations. Since the appropriate
role of ground truth is to provide information to interprete
the acoustic data, it can be most effective when the ground
truth efforts are directed to answer specific questions asked
by the acoustic (oata. Correlations between acoustic
observat ions are always best when they are synoptic. It
follows that the information value of the ground truth data
is greatest. when it can be directly compared to the synoptic
acoustic data. Whenever possible I use a two boat operation:
an acous-ic boat followed by a direct capture boat.
Finally, it should be apparent that a single ground
truth technique may not be adequate. If the acoustic data
are used to direct ground truth efforts, it follows that the
characteristics of the targets and the habitat that are
identified by the hydroacoustics should be used to determine
the best techniques to sample those targets in that habitat.
18
In many circumstances, that will require multiple ground
truth techniques.
Acknowledgments
I am grateful to many individuals and agencies for support of the studies summarized here. Initial preparationof this manuscript was supported by the Washington Sea Grant Program. I am indebted to the U.S. Agency for International Development (AID) for the funds for its completion and publication.
Literature Cited
Anon. 1984. Prudhoe Bay Waterflood Project Fish ManagementProgram 1983. Final Report, BioSonics Inc., Seattle.
Jackson, D.R. and G.L. Thomas 1979. Acoustic measurements of fish schools using array phase information. Proc. Oceans 79:59 (abstract).
Marino, D.A. 1987. An investigation of the summertime spatial and distribution behavior of kokanee (Oncorhynchus nerka) in Northwest lakes using dual beam hydroacoustic techniaues. Paper # 83, Internat. Symp. on Fish. Acoustics, Seattle.
McClain, J. 1987. A comparison of cohort proportions obtained by dual beam techniques and by tow netting in a sockeye salmon nursery lake. Paper # 85, Internat. Symp. on Fish. Acoustics, Seattle.
Peterson, G.A. 1983. A pilot study to evaluate the winter fishery biology of Pool 18, Upper Mississippi River. Final Contract Report, U.S. Fish and Wildlife Service, Rock Island Ecological Services Field Office. 13p + 4 Appendices.
Robinson, D.G. and W.F. Barraclough 1978. Population estimates of sockeye salmon in a fertilized oligothrophic lake. J. Fish. Res. Bd., Canada 35:851-860.
Thomas, G.L. 1982. Complementary hydroacoustic and net sampling techniques for determining the vulnerabilityof fish to power plant entrapment. Paper # 102, Symp.
19
on Fish. Acoustics, Bergen, Norway, 12p.
Thomas, G.L., R.E. Thorne, W.C. Acker, T.B. Staples, A.S. Kolok, L. Johnson, K. Miller, J. Yuge, B. Kulik and S. Shiba 1980. The effectiveness of a velocity cap and decreased flow in reducing fish entrapment.Final Contract Report, FRI-UW 8027, Univ. Wash. Fish Res. Inst. 22p + 9 Appendices.
Thorne, R.E. 1970. Investigations into the integrator for measuring pelagic fish aThesis, Univ. Wash., ll7pp.
use bunda
of an nce.
echo Ph.D.
Thorne, R.E. 1971. Investigations into the relation between integrated echo voltage and fish density. J. Fish. Res. Bd., Canada 28:1269-73.
Thorne, R.E. 1973. Acoustic assessment of hake, 1969-1973. Proc. IEEE' Symp. on Eng. in the Ocean Environ., 249-252.
Thorne, R.E. 1977. Acoustic assessment of hake and herringstocks in Puget Sound, Washington, and southeastern Alaska. ICES Rap. et Proc.-v. 170:265-278.
Thorne, R.E. 1979. Hydroacoustic estimates of adult sockeyesalmon in Lake Washington, 1972-1975. J. Fish. Res. Bd. Canada 36:1145-49.
Thorne, R.E. 1980. Application of stationary hydroacoustic systems for studies of fish abundance and behavior. Proc. Oceans 80:381-385.
Thorne, R.E. 1983. Assessment of population abundance byhydroacoustics. J. Biol. Ocean. 2:253-262.
Thorne, R.E. 1983b. Application of hydroacoustic assessment techniques Lo three lakes with contrasting fish distributions. FAO Fish. Rep. 300:269-277.
Thorne, R.E., J.E. Reeves and A.E. Millikan 1971. Estimation of the Pacific hake (Merluciusproductus) population in Port Susan, Washington, using an echo integrator. J. Fish. Res. Bd., Canada 28:1175-84.
Thorne, R.E., R. Trumble, N. Lemberg and D. Blankenbeckler 1983. Hydroacoustic assessment and management of herring fisheries in Washington and southeastern Alaska. FAO Fish. Rep. 300:217-222.
Thorne, R.E. and G.L. Thomas 1984. Recent applications of hydroacoustics to assessment of limnetic fish abundance and behavior. Proc. NALMS 1983 Internat. Symp. on Lake and Reservoir Management. USEPA
20
400/5/84-001, pp. 305-309.
Trumble, R., R. Thorne and N. Lemberg 1982. The Strait of Georgia herring fishery: A case history of timely management aided by hydroacoustic surveys. Fish. Bull. 80:381-388.
List of Tables
1. Spearman rank correlation coefficients between nightly mean larpera catch per set and acoustic estimates of densityfor queenfish, white croaker, north)ern anchovy and all species combined, Huntington Beach 1979-80.
2. Minimum and maximum lengths (mm) of various species caughtby gillnet, trawl and electrofishing, Mississippi River, Pool i8.
List of Figures
1. Comparison of actual catches of sockeye salmon in trawls with predicted catches based on hydroacoustic estimates of density, iLake 'Tustumena, Alaska.
2. Comparison of acoustic estimates of abundance and averagetrawl net catches of hake in Port Susan, Washington, 1969-1 973.
3. Comparison of acoustic echogram records and subsequent lampara net catches.
4. The vertical distribution of white surfperch and walleyesurfperch from vertical gill nets compared with The depthincrement of hydroacoustic observations.
5. Echogram from transect across cove rotenone area and into limnetic area of Monticello Reservoir, Texas.
6. Echogram illugftrating fish attraction to lights, Gulf of Nicoya, Costa Rica.
7. Biomass estimates of adult roe-herring in the Strait of Georgia, 1977, from various assessment techniques.
8. Cartoon illustrating principle that it is easier to catch organisms when you can see them.
(%,
1. Spearman rank correlation coefficients between nightly mean lampera catch per set and acoustic estimates of density for queenfish, white croaker, northern anchovy and all species combined, Huntington Beach 1979-80.
r n Significances
All species 0.5055 32 0.002
Queenfish 0.3221 32 0.018
White croaker 0.6328 32 0.001
Anchovy 0.3867 32 0.015
2. Minimum and maximum lengths (mm) of varicus species caughtby gillnet, trawl and electrofishing, Mississippi River, Pool 18.
Gillnet Trawl .1ectrofishing Species Min fin Max Min -Max Max
I. Comparison of actual catches of sockeye salmon in trawls with predicted catches based on hydroacoustic estimates of density, Lake Tustumena, Alaska.
1225 -1969,-"
A ;20
0 -. 1971 =15 - 1972..
CE10 ,,1970e
1973
- - - - I I 1 100 200 300 400 500 600
Average Not Catch (Ib)
2. Comparison of acoustic estimates of abundance and averagetrawl net catches of hake in Port Susan, Washington, 1969-1973.
'20
100
80
60
z0t
ki 40
204. see 0 1.3 Small wqau ' L.. e
School Scrooi Sc":: NUMBER OF TARGETS (ECHOGRAPMI
OSChoo of lwu.Sfl oflr'h0y th(al #COQSCI Ml'S troUgP' mjeptI Oes oll mo ,-gt Id-wlojr &crOol m.lsd e'lfrqly by iomporo n,"erl
Fig. 3. Comparison of icoustic echogram records and subsequent lampara
4. The vertical distribution of white surfperch and walleye surfperch from vertical gill nets compared with the depth increment of hydroacoustic observations.
p1
t- ACHAfi-0 CC
-- 5-- -- lO .. .. ...........
and intoFig. 5. Echogram from transect across cove rotenone area
1innetic area of Monticello Reservoir, Texas.
, , •c~" ' "i
VN
Fig. 6. Echogran illustrating fish attraction to 1lights, Gui f of