Editor's note: The authors of the following essay have been involved, directly or indirectly, in processes to plan a system of no-take marine reserves within the Channel Islands National Marine Sanctuary (MPA News 4:6) in the US. Michael Robinson is a doctoral student in geography at the University of California at Santa Barbara (UCSB). Chris Miller is a commercial fisherman and vice president of the California Lobster and Trap Fisherman's Association. Chris Hoeflinger is a commercial fisherman and fishery issues organizer. Barbara Walker is a geographer with the National Center for Geographic Information and Analysis at UCSB.
By Michael Robinson, Chris Miller, Chris Hoeflinger, and Barbara Walker
In October 2002, the California Fish and Game Commission adopted a regulatory action for state waters of the multiple-use, 4292-km2 Channel Islands National Marine Sanctuary (CINMS) in the US. The action closed approximately 10% of CINMS to all or most fishing. Following this, a related process was initiated to expand the reserve system to federal waters of CINMS, with the possibility of closing an additional 15% of the sanctuary. The latter process is ongoing.
To determine optimal locations for the closures in state waters, planners analyzed a variety of data layers in a geographic information system (GIS) decision-making tool. They mapped alternative MPA systems to examine theoretical impacts, with each system based on biodiversity and/or economic criteria. In an attempt by planners to minimize the economic impact of potential closures, local stakeholders were asked to participate in two surveys designed to assign values to fishing areas and habitats:
A value distribution survey collected economic data via a grid of 3000 1x1 square mile boxes superimposed over an area map. The grid was displayed on 8.5-inch by 11-inch pages, which allowed for little detail and did not include bathymetry or substrate information, both significant factors in determining fishing and habitat locations. Economic data were collected for target species that together comprised over 99% of the ex-vessel value of the 1999 CINMS commercial catch. Each fisherman was allocated 50 X's to be placed over his best area, using the 1x1 mile grid system. A value was then assigned to each block by comparing selected areas to confidential landing receipt information collected in standard 10x10 mile grids, which participants had released to a social scientist. This scientist assigned a value to each 1x1 block for each participant, based on the participant's ten-year average revenue generated in corresponding 10x10 block landing receipts.
A Traditional Ecological Knowledge (TEK) survey asked a variety of stakeholders - including commercial and recreational fishermen, marine tour operators, and recreational users - to identify areas of good habitat for specific species or marine activities. This information was combined to create an aggregate "use" map, which identified areas of good habitat and, consequently, fishing area for each species.
This was the first exposure to GIS mapping for most fishermen involved. It appears that many fishermen did not understand how the GIS tool would work, and either deliberately misrepresented their best fishing holes (to hide them from government and competitors) or unintentionally did so due to the grid system's inability to capture fine detail. This misrepresentation complicated the ground-truthing using the fish landing receipts. In addition, in the value distribution survey, many fishermen struggled when they were allocated only 50 X's for an area containing 3000 1x1 mile boxes. Lastly, the TEK data were consistently termed "anecdotal" in the decision-making process, and, from our understanding, were not directly incorporated in the GIS decision-making tool. The combination of these factors resulted in official economic impact data that, at least according to two primary commercial fisherman advocates, fell short of the intended purpose to calculate realistic financial impacts of MPA scenarios at CINMS.
In light of the above, the following are key factors that should be incorporated in the federal MPA consideration process and future MPA planning efforts:
Include bathymetric and substrate information in economic and habitat surveys, since this information plays a significant role in fish location and distribution.
Combine ecological and economic considerations in one process. Separation of these processes prevents planners from considering economic impacts and fishermen from understanding ecological considerations.
Use large maps to view bathymetric information, habitat information, and fishing effort - for example, one island per 11-inch by 17-inch map.
Update and ground-truth economic data by removing any 1x1 mile boxes that do not contain suitable habitat for target species. The map used in the economic process overgeneralized habitat, resulting in misstatement of many catch locations. Thus, much of the economic data appeared in wrong locations. Admittedly, some of this was due to misrepresentation by some fishermen, but it was also due to the difficulty of identifying proper squares on such a small map without bathymetry and substrate information. Economic impact analysis and final MPA placement were based on these incorrect catch locations.
Examine overlap of existing regulations and no-fish areas: California conservation zone, trawl exclusion zone, areas closed to gill-net fishing, etc. Identify redundancies in regulations, simplify monitoring, and determine the best locations for reserves.
Make the GIS software, data sets, and a GIS technician available to all stakeholders. This allows stakeholders to become an active part of the decision-making process. As a result, they are less likely to feel that the technology is being used against them.
For more information
Barbara Walker, National Center for Geographic Information and Analysis, Institute for Social, Behavioral, and Economic Research, University of California, Santa Barbara, CA 93106, USA. Tel: +1 805 893 3576; E-mail: blewalker [at] isber.ucsb.edu