APPLICATION OF INFRARED AND OCEAN COLOR SATELLITE DATA IN PELAGIC FISHERIES ASSESSMENT AND MANAGEMENT IN THE SOUTHEASTERN UNITED STATES

1
APPLICATION OF INFRARED AND OCEAN COLOR SATELLITE
DATA IN PELAGIC FISHERIES ASSESSMENT AND
MANAGEMENT IN THE SOUTHEASTERN UNITED STATES
Mitchell A. Roffer, Matthew A. Upton, Gregory
J. Gawlikowski, Daniel C. Westhaver Roffers
Ocean Fishing Forecasting Service, Inc. West
Melbourne, Florida, U.S.A. (WWW.ROFFS.COM) Frank
Muller-Karger School For Marine Science
Technology, University of Massachusetts-Dartmouth,
New Bedford, MA Arthur J. Mariano Rosenstiel
School For Marine Atmospheric Science,
University of Miami, Miami, Fl Donald J.
Hammond, Cooperative Science Services, LLC,
Charleston, SC
2008 Carbon Cycle Ecosystems Joint Science
Workshop, Biodiversity and Ecological Forecasting
Team Meeting, Friday May 02, 2008
2
ACKNOWLEDGEMENTS

• Chuanmin Hu - Brock Murch - Judd Taylor
• USF_IMaRS, St. Petersburg, Fl
• Barry Matz - Evan Matz Forms Processing Inc.
  Miami, Fl
• Charles Tucker - College of Charleston,
  Charleston, SC
• SUPPORT
• NASA Grant NNG04GF136
• Interdisciplinary Science Program
• Woody Turner - Lawrence Friedl
• South Carolina Department Of Natural Resources,
  Charleston, SC

3
INTRODUCTIONA CRITICAL ISSUE FOR FISHERIES
MANAGEMENT

• WHAT AFFECTS APPARENT CHANGES IN STOCK SIZE
• FISHING AND NATURAL MORTALITY?
• CHANGES IN CATCHABILITY (q) and ENVIRONMENTAL
  VARIABILITY?
• ARE GOOD CATCH RATES
• HIGH ABUNDANCE?
• INCREASED AVAILABILITY VULNERABILTY?
• ARE BAD CATCH RATES
• DECLINE IN ABUNDANCE?
• CHANGES IN AVAILABILITY / VULNERABILITY?

4
PROJECT GOALSSTEPS NECESSARY TO REACH ULTIMATE
GOAL

• DEVELOP DECISION SUPPORT TOOLS FOR RESOURCE
  MANAGERS POLICY MAKERS
• INCLUDING TECHNOLOGY TRANSFER
• DEVELOP FUNCTIONALLY BASED, STATISTICALLY
  SIGNIFICANT RELATIONSHIPS BETWEEN THE FISH AND
  THEIR ENVIRONMENT
• DEVELOP AUTOMATED TOOLS THAT ONE CAN USE WITH
  LARGE DATA BASES FOR CLIMATE RESEARCH

5
SPECIES

• DOLPHIN FISH Coryphaena hippurus

HIGHLY MIGRATORY OCEANIC Florida Keys to Rhode
Island Florida Keys to Gulf of Mexico North
Carolina to Azores Massachusetts to Bahamas
6
(No Transcript)
7
(No Transcript)
8
DYNAMIC OCEAN
9
FISH DO NOT RESPOND TO ARITHMETICALLY CALCULATED,
HIGHLY SMOOTHED MEAN CONDITIONS
CONDITIONS NECESSITATED THAT DAILY CATCH DATA AND
HIGH RESOLUTION MEASUREMENTS BE USED
10
WORKING HYPOTHESIS
11
DATA SOURCES.C.D.N.R. RECREATIONAL MAHI TAGGING
2002-2005
Manipulated With ESRI-ArcGIS
12
INDICES OF ABUNDANCE

• CATCH PER UNIT OF EFFORT
• CPUE WITH VOLUNTEER RECREATIONAL TAGGING IS AN
  ESTIMATE OF EXCESS CATCH OF FISH PER DAY
• FISHERS ONLY TAGGED WHEN THEY HAD ENOUGH FOR
  THEMSELVES.
• LEGAL BAG LIMIT OF 10 PER PERSON/60 PER BOAT
• EXCESS IS VARIABLE, BUT HIGH CATCH RATES
  INTERPRETED AS VERY HIGH AND LOW TAG RATES AS LOW
  CATCH RATES
• ONLY TAGGED FISH REPORTED.
• ACCURATE LOCATION DATA

13
AS ADVERTISED A REVIEWLESSONS LEARNED -
RE-LEARNED

• Useful information that were derived from
  satellites on a daily basis was supplied to
  resource managers to better understand changes in
  catch.
• Dolphin fish were shown to be statistically
  related to oceanic fronts including the Gulf
  Stream.
• Daily high resolution measurements must be used
  to evaluate changes in the distribution and
  relative apparent abundance of fish.
• Gulf Stream frontal boundaries and other nearby
  fronts along with SST are important for defining
  and characterizing the essential fish habitat.
• Fish caught over 100m depths compared with 200m.

14
REVIEW CONTINUED

• Inshore location of the Gulf Stream front was
  related to higher catches.
• Statistically significant relationships between
  the fish and eddy characteristics were elusive.
• Except number of eddy features and total catch.
• Infrared data were more useful than visible
  (ocean color) due to repeat coverage to reduce
  the effect of clouds and other atmospheric
  interference.
• Banding issues with Terra-MODIS exacerbated issue
• For defining frontal boundaries manual
  (non-automated), human analysis techniques were
  more useful than automated techniques, especially
  when significant atmospheric interference exists.
• Limitations of fishery dependent data were
  apparent, but did not preclude research successes.

SHOW SOME RESEARCH RESULTS
15
MONTHLY TAG RELEASE
16
DOLPHIN FISH AND SST
MEAN 25.1C MIN. 20.5C MAX. 29.5C
17
DISTANCE OF FISH TO NEAREST FRONT

• 50 - 2.6 km 75 - 5.0 km 85 - 7.6 km

18
DISTANCE TO WESTERN EDGE OF GULF STREAM
50 - 9.4 km 75 - 20.7 km 85 - 23.2 km
19
DECISION SUPPORTMANAGERS FROM SC D.N.R CALLED

• WAS THE APPARENT DECLINE IN THE TOTAL NUMBER OF
  REPORTED CAPTURED FISH AND TAGGED FISH IN MAY
  2005 AN INDICATOR OF DECLINING STOCK ?
• WE WERE NOT READY FOR THIS.
• SOMETIMES WE MUST AS SCIENTISTS

20
CATCH (TAG RELEASE) PATTERNS 2003-2005
21
COMBINED
22
APRIL 18-21, 2004 vs. MAY 05-09, 2005
23
BUT WE LEARNED DIFFERENTLY AFTER COMPLETING OUR
ANALYSES
24
EDDY COMPARISONMORE EDDIES IN 2005 THAN 2004
25
2004 VS. 2005 AVAILABILITY NOT ABUNDANCE

• FAVORABLE FRONTAL BOUNDARY CONDITIONS CLOSER TO
  COAST IN 2004
• FRONTS MORE STABLE IN 2004
• WEATHER BETTER IN 2004
• ANGLERS COULD SPEND MORE TIME CATCHING FISH,
  RATHER THAN SEARCHING RANDOMLY FOR PRODUCTIVE
  HABITATS IN 2004 COMPARED WITH 2005.
• THE FIRST ONE TO THE FISH WINS

26
EDDY CLASSIFICATION CHALLENGES

• Sized By Class
• Some Change Size
• Moving Targets

LOOKING FOR COLLABORATORS
27
Technology TransferDECISION SUPPORT

• SOUTH CAROLINA DEPT. NATURAL RESOURCES. - DOLPHIN
  FISH
• NC GA
• NOAA FISHERY COUNCILS INTERESTED
• FLORIDA OCEAN AND COASTAL COUNCIL
• FL DEP FWRI FISH, RED TIDE, AQUACULTURE
• NORTH CAROLINA DEPT. HEALTH
• RED TIDE CIRCULATION ANALYSES
• NOAA_NMFS SEFSC
• ATLANTIC BLUEFIN TUNA POPULATION ASSESSMENT NASA
  NNX08AL06
• OTHER FISH LARVAE RESEARCH
• BLUEFIN, BILLFISHES, OTHERS- UNIV. S. MISS.

28
BLUEFIN TUNA POPULATION ASSESSMENT ADDAPTIVE
SAMPLING TO POPULATION ASSESSMENT

• TOOLS MERGING SST AND CHLOROPHYLL FOR HABITAT
  SELECTION

29
ADAPTIVE SAMPLING

• FRONTAL ANALYSES
• 30 APRIL 2008

30
MOVIES FOR MANAGERSMODIS OCEAN COLOR
------------ MODIS/AVHRR INFRARED

• UTILITY NOT NECESSARILY PRETTY

31
THANK YOU QUESTIONS
32
SLIDES FOR QUESTIONS
33
NUMBER OF EDDY FEATURES CPUE
34
WORK CONTINUES

• 2003-2005 JUST A START
• NEED LONGER TIME SERIES STUDIES
• IMPROVE AUTOMATIC FRONTAL DETECTION - WEAKER
  GRADIENTS?
• CONSIDER CHLOROPHYLL FRONTS

35
MANAGERS NEED

• Need to be trained on how to use the satellite
  data
• Need Desktop Visualization And Manipulation Tools
  To Facilitate Routine Use Of Satellite Other
  Environmental Data.
• Easy To Use !
• Need to acquire value added - interpreted
  satellite derived oceanographic data on the
  appropriate temporal and spatial scales.
• e.g. ocean frontal analyses

36
LIMITATIONS OF DATA

• NEED LONGER TIME SERIES TO DERIVE MORE
  STATISTICALLY SIGNIFICANT RELATIONSHIPS
• MORE FISH CATCH DATA
• Expensive

37
FRONTAL DETECTION
38
EFFECT OF ATMOSPHERICINTERFERENCE
39
(No Transcript)