Geospatial Applications in Aquaculture and Fisheries Management
(8/13/2019 12:00:00 AM)
Abstract
Geospatial
applications can contribute to fish distribution, abundance, and movement
analyses. They are particularly useful
for spatial management strategies for fisheries. However, their feasibility must be assessed
to determine utility. Geospatial
applications cannot address all management questions and often their large data
demands are significant limitations.
“Managing
fisheries is hard: It’s like managing a forest, in which the trees are
invisible and keep moving around.” -Professor John Shepherd, University of
Southampton
Spatial
Management Strategies
Spatial management
strategies are a tool to help optimize fisheries management and are often used
as a complement to more traditional harvest management techniques such as catch
quotas (e.g., total allowable catch [TAC]), fishing effort limits (e.g.,
limited number of boats, gear, or trips, and size limits on fish harvested
(e.g., minimum length). Spatial
management strategies include fishery designation zones (e.g., no trawl areas),
closed seasons for particular locations, and closed areas (e.g., marine
protected areas [MPAs]). Key issues to
consider for successful implementation of spatial management strategies
include: (1) all relevant jurisdictions should be engaged; (2) local
communities should be involved; (3) enforcement must be feasible and
functional.
 Spatial strategies
can be optimized, and geospatial applications can be used to assist in this
process. They can help estimate fish
distribution, abundance, and movement.
Spatial analyses for fish populations can range from qualitative, such
as visualizations, basic mapping, and multi-parameter overlays, to
quantitative, such as georeferenced population dynamics and assessment and
space-based fisheries management and prediction.
Feasibility questions for spatial management of fisheries:
● What
are the key issues relating to the management and further development of the
fisheries and/or aquaculture sector?
● Which
of these issues are fully or partially spatial?
● What
geographic area should be covered?
● What
georeferenced data do you have? What
data are needed?
● To
what extent will GIS satisfy information needs?
● Would
funding be available for any longer-term GIS work?
● Are
there better ways of achieving the information that is needed?
Fish
Distribution
Analyses of fish
distribution ask the (often not so) simple question of where fish are
located. These studies can examine fish
across a geographic region (e.g., a species range) or within a given
environment (e.g., zones of a lake).
Geospatial tools can help define fish distribution. Habitat modeling can visualize spatial
distributions of quantity and quality of fish habitat (for more information, see habitat modeling section). Species distribution modeling can produce
predictive maps of where species are likely to occur and not likely to occur
based on field observations. FishTail (https://ccviewer.wim.usgs.gov/Fishtail/) is an example
tool that examines fish distribution projections with climate change. It is a decision-support mapper that
integrates land use, habitat fragmentation, water quality, and future climate
conditions to provide risk estimates of changes in fish habitat for stream fish
in the Northeastern United States.
Feasibility questions for geospatial applications for fish
distribution:
● Which
species do you manage with uncertain distributions?
● What
data do you need? (e.g., primary data or secondary data)
● Are
there implications for changing conditions? (e.g., climate, dams, land use)
Fish
Abundance
Analyses
of fish distribution examine how many fish are present in a given spatial
context. Often, these studies reference
the density of fish within a given geographic area. Geospatial tools can provide spatial
interpolations to examine correlations between fish abundance and environmental
variables (e.g., habitat types). Dueri
et al. 2014,
for example, used the relationship between climate and skipjack tuna abundance
and distribution to project changes in biomass of the species with climate
change.
Feasibility questions for geospatial applications for fish abundance:
● How
can geospatial data be used to examine abundance estimates?
● What
data do you need? (e.g., primary data or secondary data)
● Why
might accounting for spatial variability improve stock management?
Fish movement
Analyses of fish
movement assess where fish are going.
These studies can track fine-scale daily behavior of fish and can
examine large-scale migrations, depending on the monitoring equipment utilized. Pop-up satellite archival tags (PSATs) record
information such as temperature, light level, oxygen levels and pressure over a
researcher defined interval of seconds to hours. After a predetermined duration, the tag
releases from the animal and, once at the surface, transmits the information to
researchers via satellite. These tags
are principally for large oceanic species because they are expensive and can be
lost due to biofouling or premature release.
Passive Integrated Transponder (PIT) tags, cannot provide environmental
information like PSATs, but use a small, injectable, radio transponder to
“barcode” fish. The fish retain this
acoustic tracking tag for life and transmit their individual identifying code
when they pass an acoustic receiver. PIT
tags are considerably cheaper than PSATs and well-placed arrays of acoustic
receivers can be used to examine movement patterns and habitat usage. The information provided from fish tags can
be analyzed to visualize timing and location of migration and important life
history events as well as identify critical locations for conservation and
protection.
Feasibility questions for geospatial applications for fish movement:
● Which
migratory species do you manage?
● What
data do you need? (e.g., primary data or secondary data)
● How
might geospatial analyses guide management? (e.g., fishing restrictions by
season or location?)
Abstract
Habitat mapping
and modeling are important for effective land and resource management. Defining
and identifying critical habitat as
well as threats to those habitats are
also important for effective management. Habitat maps may be created via
digitizing and field work or modeling within a GIS. The basic modeling process
comprises six steps: 1) Define objectives and purpose, 2) State assumptions, 3)
Identify model variables, 4) Locate GIS data representing the model variables
at the desired scale, 5) Implement the model, 6) Evaluate model results. A
well-designed model and GIS tools allow for re-running of models with adjusted
inputs to reflect changing conditions or management objectives.
Why Map Habitat?
Knowledge of
surrounding habitats and the species that live within them is critical for
effective land and resource management and provides a basis for ecological
research. Understanding where different ecological systems are and how they
function is an important guide for intelligent management decisions that
promote sustainability. Figure 6 diagrams the interplay between habitat mapping
for management and for ecological understanding.
Figure 6. Comparison of reasons for habitat mapping for management and
research.
Critical Habitats and Threats
Crucial or critical habitat are
those habitats known to be important for either ecological or economic reasons.
In many cases, economic reasons may depend on underlying ecological health but
whether economic or ecological, identifying critical habitat helps prioritize
management decisions.
The United
States Fish and Wildlife Service maintains geospatial data of critical habitat
(https://ecos.fws.gov/ecp/report/table/critical-habitat.html). These data
are often used in a variety of project planning and geospatial modeling.
In addition to
mapping habitats, it is important to also map threats to habitats; which are
also threats to the species using the habitats. Maps of threats compared in a
GIS to maps of habitat--especially critical habitat--provide a geographic focus
for management decisions by showing where habitats are under threat. Examples
of threats to species and habitats may include industrial areas, resource
extraction, transportation networks and urbanization. Development of a holistic
threat map may require the creation and mapping a host of features on the
landscape.
Creating Habitat Maps
Digitizing
A commonly used
method of habitat mapping is simply drawing in a GIS on top of imagery, or digitizing. A skilled person with
knowledge of the landscape may be able to create GIS data just by looking at
imagery and zooming in and out and identifying difference in the imagery that
equates to changes in habitat. Existing paper maps with valuable geodata can be
“heads-up digitized” into a GIS. A GIS specialist simply copies by hand and eye
the locations, line work and areas by comparing the hard copy map to similar
maps in the GIS and using tools in the GIS to recreate the features. Existing
maps may also be digitized by being scanned or photographed to be loaded into a
GIS. Spatial data on scanned maps can then be traced using data creation tools
in a GIS. Any of the digitization processes is usually accompanied by field
work to “ground truth” habitat areas identified in imagery.
Modeling Habitat
If direct
mapping of habitat via field work and digitizing is not possible, a predictive
model of areas likely to contain certain habitats can be built in a GIS. The
modeling process includes the following steps:
1. Define
objectives and purpose
2. State
assumptions
3. Identify
model variables
4. Locate
GIS data representing the model variables at the desired scale
5. Implement
the model
6. Evaluate
model results
Habitat
Suitability Index (HSI) models were developed by the U.S. Fish and Wildlife
Service (FWS) in the early 1980’s as part of the Habitat Evaluation Program.
Handbooks and policy developed by the FWS is available at https://www.fws.gov/policy/ESMindex.html.
Example -- Fish
Habitat Suitability Modeling in Florida Estuaries
https://www.academia.edu/27764855/Spatial_Modeling_of_Fish_Habitat_Suitability_In_Florida_Estuaries
Scientists at
the Florida Fish & Wildlife Research Institute developed spatial habitat
suitability models (HSM) using GIS to predict spatial distributions and
relative abundance of fish species at various life stages in two major Florida
estuaries. Data from sample locations were interpolated to produce seasonal
habitat maps for dissolved oxygen, salinity, temperature. Model inputs for
bathymetry and bottom type were also produced.
Fish catch data
were compared to the different habitat parameters to build a predictive model
across an estuary. The model algorithm was then applied to a second estuary and
catch data from there were used to test the model. This process of transferring
models between catch data from two estuaries refined the model and increased
understanding of fish habitat for several species. The refined model
information on preferable fish habitat conditions allows the model to be
applied in regions for which catch data are not available.
Well-designed
models and associated GIS processes can be used to test multiple scenarios.
Specific input layers and the underlying algorithm used in the model can be
adjusted and the model re-executed to derive results for different expected
conditions. In the case of climate change, habitat models are often re-run with
updated, predicted temperature conditions. Species distribution changes
expected as a result of climate change can be produced by re-modeling a current
distribution map but with a predicted temperature regime layer. Changing
management objective can also be reflected in re-modeling to provide guidance
on proposed management actions.
Any GIS modeling
process includes five major steps, define the objectives and purpose, identify
the model variables, locate GIS data representing the model variables at the
desired scale, implement the model, and evaluate the model results. The figure below lists some aquaculture
suitability criteria variables:
Three different
papers were presented that demonstrate aquaculture site suitability, two were
located in Vietnam and the last one is in Tenerife Canary Islands of the north
east coast of Africa.
Example 1: GIS for land
evaluation for shrimp farming in Haiphong of Vietnam
Abstract
This
study was conducted to identify appropriate sites for shrimp farming
development in Haiphong province of Vietnam using geographical information
systems (GIS). Thirteen base layers (thematic maps) were grouped into four main
land use requisites for aquaculture, namely, (1) potential for pond
construction (slope, land use type, soil thickness, elevation), (2) soil
quality (soil type, soil texture, soil pH), (3) water availability (distance to
sea, and water source), and (4) infrastructure and socio-economic status
(population density, distance to roads, local markets, and hatcheries). A
constraint layer was used to exclude areas from suitability maps that were not
allowed to implement shrimp farming. A series of GIS models was developed to
identify and prioritize the most suitable areas for shrimp farming.
This
study shows that the land evaluation model is useful for identifying suitable
areas for shrimp farming and for allocating land for efficient income
generation, effective conservation, and sustainable land management. It was
estimated that about 31% (2604 ha) of the total land area (8281 ha) in Haiphong
was highly suitable for shrimp farming. Since existing shrimp farms cover only
1690 ha of land in the study area, the potential for expanding shrimp farms
should take into consideration further political and environmental issues.
Example 2: Investigation of
a novel approach for aquaculture site selection
Abstract
This
study investigated the potential use of two “species distribution models”
(SDMs), Mahalanobis Typicality and Maxent, for aquaculture site selection. SDMs
are used in ecological studies to predict the spatial distribution of species
based on analysis of conditions at locations of known presence or absence. Here
the input points are aquaculture sites, rather than species occurrence, thus
the models evaluate the parameters at the sites and identify similar areas across
the rest of the study area. This is a novel approach that avoids the need for
data reclassification and weighting which can be a source of conflict and
uncertainty within the commonly used multi-criteria evaluation (MCE) technique.
Using pangasius culture in the Mekong Delta, Vietnam, as a case study,
Mahalanobis Typicality and Maxent SDMs were evaluated against two models
developed using the MCE approach. Mahalanobis Typicality and Maxent assess
suitability based on similarity to existing farms, while the MCE approach
assesses suitability using optimal values for culture. Mahalanobis Typicality
considers the variables to have equal importance whereas Maxent analyses the
variables to determine those which influence the distribution of the input
data. All of the models indicate there are suitable areas for culture along the
two main channels of the Mekong River which are currently used to farm
pangasius and also inland in the north and east of the study area. The results
show the Mahalanobis Typicality model had more high scoring areas and greater
overall similarity than Maxent to the MCE outputs, suggesting, for this case
study, it was the most appropriate SDM for aquaculture site selection. With
suitable input data, a combined SDM and MCE model would overcome limitations of
the individual approaches, allowing more robust planning and management
decisions for aquaculture, other stakeholders and the environment.
Example 3: Geographical
information systems-based models for offshore floating marine fish cage
aquaculture site selection in Tenerife, Canary Islands
Abstract
The present study focuses on the development
of a standard methodology for selection of suitable sites for offshore
(exposed) marine fish-cage farming (floating cages) of seabream (Sparus aurata) and seabass (Dicentrarchus labrax) in an
island environment, using Tenerife as an example. Site selection is a key
factor in any aquaculture operation, affecting both success and sustainability
and can solve conflicts between different activities, making a rational use of
the coastal space. Site selection was achieved by using geographical
information systems (GIS)-based models and related technology to support the
decision-making process. The framework for spatial multi-criteria decision
analysis used in this study began with a recognition and definition of the
decision problem. Subsequently, 31 production functions (factors and
constraints) were identified, defined and subdivided into eight submodels.
These were then integrated into a GIS database in the form of thematic layers
and later scored for standardization. At this stage, the database was verified
by field sampling to establish the quality of data used. The decision maker’s
preferences were incorporated into the decision model by assigning weights of
relative importance to the evaluation under consideration. These, together with
the thematic layers, were incorporated using multi-criteria evaluation techniques
and simple overlays to provide an overall assessment of possible alternatives.
The integration, manipulation and presentation of the results by means of
GIS-based models in this sequential and logical flow of steps proved to be very
effective for helping the decision-making process of site selection. Tenerife
has very favorable environmental conditions for culture of marine fish and
there are no totally unsuitable sites for cage farming identified in this
study. On the other hand, there are few very suitable sites (high scores)
either, principally due to the heavy use of the coastline and the conflicts
between different users. From the 228 km2 of available area for siting cages in
the coastal regions with depth less than 50m, the total area suitable for
siting cages (scores 6-8) was 37 km2. There are only 0.51km2 of very suitable
areas (score 8) and approximately 5.37 km2 of suitable (score 7), most of these
being located in the southeast of the island. These relatively small areas of
suitability should be put into the context of the wider use of the coastal
environment around Tenerife.
Ø Watershed Management and
Monitoring
Abstract
Watershed
management is a comprehensive strategy to protect and improve water quality and
natural resources within a watershed.
All activities within a watershed affect the watershed’s natural
resources and water quality. Geospatial
applications can contribute to watershed management and monitoring.
Watershed Basics
A watershed is
the area of land where the water that drains off it goes to a common
outlet. Depending on scale, both basin
(large-scale) and catchment (small-scale) are terms often used interchangeably
with watershed. In general, a basin
references a large river system and a catchment references a small stream. Watershed management is the land use and
water use practices used to protect and improve the quality of water and other
natural resources within a watershed by managing in a comprehensive manner. Ultimately, the goal of watershed management
is sustainable development and management to preserve natural resource
functions and biodiversity for ecological health, economic prosperity, and
social security.
Importance of Watershed Management
Watershed
management is important because all activities within a watershed affect the
watershed’s natural resources and water quality. These activities include land development,
runoff from urban areas, and agricultural activities. Management within these systems can control,
reduce, and even eliminate pollution sources from these activities within a
watershed. Watershed management is often
an exercise in political diplomacy as watershed boundaries do not coincide with
political boundaries. But, for that
reason, watershed management efforts are particularly important to ensure
effective implementation of restoration and conservation measures. What happens in the headwaters of a river
affects the delta. In the Mekong River,
for example, 16% of the mean annual flow is from China, 2% from Myanmar, 35%
from Lao PDR, 18% from Cambodia, 18% from Thailand, and 11% from Viet Nam. However, the majority of inputs from China,
Myanmar, Lao PDR, Cambodia, and Thailand, all end up down in the delta region
of Viet Nam.
How Watershed Management Works
Watersheds are
most effectively implemented as a participatory process. The most productive examples include local
partnerships of, for example, residents, landowners, governments, developers,
and agricultural users. Often,
negotiations between these different stakeholders with diverse interests is
necessary. And effective implementation
should be adaptive and capable of responding to feedback from monitoring to
ensure progress towards meeting the intended goals. Monitoring often results in iterative
adjustments to guidelines and policies to make greater strides in achieving
watershed management goals.
Priorities for
management action can include infrastructure improvements, reductions in
impervious surfaces, improved agricultural practices, habitat restoration,
improved waste management, and increased riparian buffers. Considering a hypothetical watershed example,
taking a watershed management approach could result in decreased sources and
transport of nutrients, sediments, and contaminants; decreased nutrients from
atmospheric deposition and wastewater discharge; increased wetlands and forest
habitat; control of water withdrawals; and promotion of sustainable fisheries
harvest. These management actions could
result in improved oxygen concentrations; fewer algal blooms and improved water
clarity; increased submerged aquatic vegetation; sustainable water
availability; and improved fish and bird populations.
Geospatial Applications of Watershed
Management
Because
watershed management is an inherently spatial process, geospatial tools can
help implement strategies effectively.
Geospatial tools can delineate and map watershed boundaries and smaller
drainages basins within a watershed.
They can inventory and map natural resources, land use and land cover,
soils (and areas of erosion), and pollution (both point and nonpoint) within a
watershed. Using this geospatial
information, the watershed management process can more appropriately identify
targets for action and effective implementation.
Chesapeake STAT is one example of a geospatial
tool used for watershed monitoring. The
Chesapeake Bay is the largest estuary in the United States. The Chesapeake Bay watershed is 64,000 miles2
and 17 million people reside within its boundaries. Consequently, it is subject to many different
nutrient and pollution sources and coordinating restoration and management
efforts across the multijurisdictional watershed is challenging. Chesapeake STAT is a tool developed by the
Chesapeake Bay Foundation to improve information sharing and decision making
within the watershed. The online portal
houses a range of environmental data time series that can be used to help track
progress towards meeting watershed restoration and management objectives.
For
more information and training in watershed management, please visit the U.S.
Environmental Protection Agency (EPA)’s online training
in watershed management. This is an
online learning program that offers self-paced training modules as a basic
introduction to watershed management including watershed ecology, watershed
change, analysis and planning, and management practices.
VIFEP (USAID
workshop)
|