Analysis of Important Bird Areas in the Bering Sea
The climate is changing in the Bering Sea and Aleutian Islands region. Residents, stakeholders, scientists, and natural resource managers are all concerned about the impacts of future climate change on important species, systems and habitats. But projections of future climate are uncertain, and different approaches have different strengths and limitations. With variation in projections, and with different impacts depending on species and systems of concern, how can stakeholders and managers realistically anticipate and plan for the impacts of climate change?
Over the past months, an interdisciplinary team of scientists has begun an effort to assess the vulnerability of key resources and ecosystem services including marine mammals, fish and commercial fisheries, terrestrial vegetation, human community sustainability and seabirds. Vulnerability assessment is a well-developed, systematic way to “synthesize and integrate scientific information, quantitative analyses, and expert-derived information in order to determine the degree to which specific resources, ecosystems, or other features of interest are susceptible to the effects of climate change” (USFS 2011). The Aleutian and Bering Climate Vulnerability Assessment (ABCVA) proposed to integrate model projections from two recent climate downscaling approaches; one from the Bering Sea Project, and one from the Spatial Tools for Arctic Mapping and Planning (STAMP). Both of these model outputs were ingested into the AOOS data system and results were shared with a team of 30 leading researchers and managers at the recent Alaska Marine Science Symposium in January. Since then, volunteer teams of expert working groups have begun to develop preliminary vulnerability assessment. Teams have identified specific information gaps and questions of greatest relevance to resource managers and other stakeholders in the region relative to projected changes in climate.
One team, focused on seabirds, has identified the need for an analytical process that would allow the the exploration of multiple climate projection data sets. This analytical process, described below, provides an ideal opportunity to evaluate the DMAC system integration capabilities within an existing project with Federal and non-Federal partners and Regional Associations. This project is a collaboration between the Aleutian and Bering Sea Islands Landscape Conservation Cooperative (ABSI LCC), the Alaska Climate Science Center and the Alaska Ocean Observing System (AOOS). It represents approximately a $150,000 investment, contributed in-kind staff time from the three organizations leading the effort, as well as substantial investments of in-kind investments from the 30 members that make up our 5 focal teams. The seabird team, which includes collaborators from Audubon Alaska and the U.S. Fish and Wildlife Service, has identified the need for a change detection analysis using 13 projected climate, and climate-derived, raster data layers within the boundaries of Audubon’s recently established Important Bird Areas (IBAs). Important Bird Areas are zones of high biological productivity and conservation priority for marine bird species extend throughout the Pacific Flyway up the coast of North America, from Mexico to the Gulf of Alaska, through the Bering Sea and into the Arctic Ocean (Smith et al. 2014). Almost 80% of the breeding seabirds within the U.S. are dependent on the Bering Sea and Aleutian Islands making IBAs within our study region especially important. This process will enable access to complex model outputs, and will create new pathways to discover and explore a broad suite of projected model variables that impact seabirds. Pairing this change detection analysis with expert knowledge on seabird ecology will strengthen the ability of managers to proactively plan for seabird monitoring and conservation.
Opportunities to replicate and adapt the workflow processes established by the IOOS System Integration Test could include any other regions where user-defined polygons and IOOS held model projects can be used to define statistical reports for IOOS resources using open-source computing resources. Other examples include using a very similar process to explore the projection space with IBAs in California using the same types of models, or marine mammal distribution areas and how they relate to projections of food sources (e.g., foraging areas associated with rookeries for northern fur seals and Steller sea lions have similarly been identified by NOAA managers). The proposed IPython Notebooks could also be used to explore projected changes in the marine ecosystem and how they relate to Essential Fish Habitat maps.
Can all the PMEL models and their corresponding variables be adequately discovered and accessed using IOOS tools?
Can we discover, access, and overlay Important Bird Area polygons (and therefore other similar layers for additional important resource areas) on PMEL climate projections for the Bering Sea?
Is metadata for projected climate data layers and Important Bird Area polygons sufficient to determine a subset of polygons desired by a query?
Can a simple set statistics (e.g., mean and standard deviation) be derived from multiple variables in each of the six models to derive the forecast variability of climate conditions through time, through the end of the model runs (2003-2040)?
Can we create a standardized matrix or other display method for output variables that allow resource experts to easily assess projected changes in climate variables, within given ranges of time, and compare projected changes across multiple coupled oceanographic and climate models?
Can we develop a set of process-specific guidelines and a standardized set of outputs for a tool that would allow researchers to address a diversity of resource management questions relative to projected changes in climate for specific zones of interest?
- Important Bird Areas (IBAs) defined by the Arctic Marine Synthesis study
- PMEL-run Model Forecasts for Bering Sea (hosted by AOOS) that are oceanographic-climate coupled models:
- ECHO-G
- CCCma
- MIROC
##Variables contained in each PMEL Model:
- Sea Water Temperature
- Small Phytoplankton Concentration
- Euphausiids Concentration
- Benthic Infauna
- Current Velocity
- Benthic Detritus Concentration
- Small Coastal Copepod Concentration
- Large Microzooplankton Concentration
- Neocalanus Concentration
- Large Phytoplankton Concentration
- Offshore Neocalanus Concentration
- Ice Phytoplankton Concentration
- Sea Ice Area Fraction
If time permits, secondary PMEL-produced models could also be tested (but results kept separate from the above)
- FORECAST
- NCEP CFS-R
- CLIVAR
Other oceanographic-only models held by AOOS that could be tested for changes in projected climate variability (but not projected biological variability) include:
- SNAP AR5 Climate Models from AR5 GFDL-CM3, RCP 6.0 (Near Surface Winds, Air Temperature)
- SNAP AR5 Climate Models from AR5 MRI-CGCM3, RCP 6.0 (Near Surface Winds, Air Temperature)
- SNAP AR5 Climate Models from AR5 IPSL-CM5A-LR, RCP 6.0 (Near Surface Winds, Air Temperature)
- SNAP AR5 Climate Models from AR5 GFDL-CM3, RCP 8.5 (Near Surface Winds, Air Temperature)
- SNAP AR5 Climate Models from AR5 MRI-CGCM3, RCP 8.5 (Near Surface Winds, Air Temperature)
- SNAP AR5 Climate Models from AR5 IPSL-CM5A-LR, RCP 8.5 (Near Surface Winds, Air Temperature)
- SNAP Projections of Sea Ice Area Fraction from CESM-CAM5, CMCC-CM, ACCESS-1, MIROC-5, and HAD-GEM2-AO