Aug 20,2018
Meeting attendees: Remi Daigle (Laval University), Anna Metaxas (Dalhousie University), Ryan Stanley (Fisheries and Oceans Canada), Gary Pardy (Fisheries and Oceans Canada)
Purpose: This meeting was scheduled to discuss the initial coordination of the Fisheries and Oceans Canada - Marxan Connect collaborative case study. The intention was to clarify specific roles, expectations, and data availability.
We first clarified roles and expectations:
- For Nov 1, the goals are to conduct a Marxan analysis using existing layers that DFO Maritimes will provide which includes connectivity. DFO practitioners of all levels (i.e. very familiar with Marxan: Gary Pardy [Oceans Biologist at Fisheries and Oceans Canada], somewhat familiar with Marxan: Marty King [MPA Network Project Lead at Fisheries and Oceans Canada], and not familiar with Marxan: possibly Maxine Westhead [Marine Protected Areas Section Head at Fisheries and Oceans Canada and others) will provide written feedback on the use of Marxan Connect as well as the results. This feedback is designed to collect constructive criticism of the website, the app, the workflow/methods, and the end results with respect to user-friendliness, clarity, completeness, suitability, and usability.
- Jennifer McGowan (The Nature Conservancy) will assist in structuring the feedback process
- This feedback is vital to the project as the objective is to make Marxan Connect broadly adopted
- Comparing the current DFO network plan with a network plan "with connectivity" is unlikely to happen by Nov 1: however, DFO practioners and decision makers are interested in such a comparison in the long term. Additional funding would be needed for this.
Issues of data availability and sharing were also discussed:
- Because of data privacy concerns, some data layers cannot be shared publicly.
- All parties agreed that it would be ideal if a subset of data could be shared publicly to be included in a tutorial for the app.
- Daigle could develop a tutorial based on the publicly shareable layers, should funds become available to do so. DFO practitioners could use this tutorial to run the "full" analysis that includes all the layers.
- The "ecological layers", a subset of the data that describes fine and coarse scale conservation objectives (e.g. habitats, species etc) may be fully shareable but this remains to be discussed internally within DFO.
- Pending internal approval for sharing with 1) the project team and 2) the public, the "ecological layers" should be shared in the first week of September.
- DFO uses Marxan with Zones for their full analysis but compatibility with Marxan Connect is unknown. DFO will share the layers used in Marxan with Zones with Daigle in the first week of September, to test compatibility with Marxan Connect.
- Stanley’s habitat suitability modelling results could be used as a resistance surface to calculate connectivity.
TASK Open
Remi Daigle16
Posted in Projects
Marxan Connect
and visible to The Public
Case Study: Fisheries and Oceans Canada - Maritimes Region
Collaborate with Fisheries and Oceans Canada - Maritimes Region to conduct case study
by Remi Daigle
Proposed project timeline:
- Marxan data transfers completed by September 14th
- Daigle, Stanley, and Metaxas to discuss specifics of connectivity data by September 21st
by Remi Daigle
Sep 18, 2018
Meeting attendees: Remi Daigle (Laval University), Anna Metaxas (Dalhousie University), Ryan Stanley (Fisheries and Ocean Canada)
Purpose: This meeting was scheduled to discuss the available connectivity data for the Fisheries and Oceans Canada - Marxan Connect collaborative case study. The intention was to identify appropriate connectivity data for the case study.
First we discussed the different types of "connectivity" that Marxan Connect can handle and their pros & cons:
- Isolation by distance (IBD) (i.e. the distance between habitat patches) has the lowest data requirements but is also likely to be the least informative.
- IBD is important, but many other factors affect movement (e.g. currents, behaviour).
- IBD would only be useful in scenarios where the habitat is patchy and no other data is available
- Isolation by resistance (IBR) (i.e. the propensity for the habitat between the habitat patches of interest to promote or prevent movement).
- Resistance to movement is typically measured using tagging etc to quantify movement rates through different habitat types
- Resistance is often estimated using habitat suitability model outputs or even species community data
- Larval dispersal models may provide more appropriate movement data, but often lack in terms of full life-cycle recruitment and survival
- Stanley's genetic data suggests that IBR may be a viable approach
- Large scale genetic discontinuities such as the "Halifax line" where the populations of various species show genetic differences on either side of the line should be treated as separate populations (i.e. separate representation and connectivity targets north and south of the boundary).
We discussed that we had to be clear about:
- Which data will be used and what assumptions are associated
- What kinds of assumptions and caveats are associated with the methods used in incorporating or even calculating connectivity
- We may want to weigh layers that have more certainty higher than layers associated with high uncertainty
- For many habitat suitability models, resolution and accuracy may be an issue. We may want to use suitability categories (i.e. top 20% is suitable, bottom %20 is not suitable)
We decided which approaches to use for connectivity in the case study:
- We will use IBR using habitat suitability for Cusk (Michelle Greenlaw - Fisheries and Oceans Canada) and Cod (Stanley).
- We will also use the stratified abundance estimates from DFO's research trawl surveys for IBR calculations since these data are trusted, available for many species, and standardized across DFO regions.
- The trawl survey data has depth limits (~300m?)
- Ellen Kenchington (Fisheries and Oceans Canada) has more habitat suitability models, mostly for corals. These may reflect historical sampling patterns more than actual coral distribution patterns. We have decided not to use these data for this phase of the case study.
Meeting attendees: Remi Daigle (Laval University), Anna Metaxas (Dalhousie University), Ryan Stanley (Fisheries and Ocean Canada)
Purpose: This meeting was scheduled to discuss the available connectivity data for the Fisheries and Oceans Canada - Marxan Connect collaborative case study. The intention was to identify appropriate connectivity data for the case study.
First we discussed the different types of "connectivity" that Marxan Connect can handle and their pros & cons:
- Isolation by distance (IBD) (i.e. the distance between habitat patches) has the lowest data requirements but is also likely to be the least informative.
- IBD is important, but many other factors affect movement (e.g. currents, behaviour).
- IBD would only be useful in scenarios where the habitat is patchy and no other data is available
- Isolation by resistance (IBR) (i.e. the propensity for the habitat between the habitat patches of interest to promote or prevent movement).
- Resistance to movement is typically measured using tagging etc to quantify movement rates through different habitat types
- Resistance is often estimated using habitat suitability model outputs or even species community data
- Larval dispersal models may provide more appropriate movement data, but often lack in terms of full life-cycle recruitment and survival
- Stanley's genetic data suggests that IBR may be a viable approach
- Large scale genetic discontinuities such as the "Halifax line" where the populations of various species show genetic differences on either side of the line should be treated as separate populations (i.e. separate representation and connectivity targets north and south of the boundary).
We discussed that we had to be clear about:
- Which data will be used and what assumptions are associated
- What kinds of assumptions and caveats are associated with the methods used in incorporating or even calculating connectivity
- We may want to weigh layers that have more certainty higher than layers associated with high uncertainty
- For many habitat suitability models, resolution and accuracy may be an issue. We may want to use suitability categories (i.e. top 20% is suitable, bottom %20 is not suitable)
We decided which approaches to use for connectivity in the case study:
- We will use IBR using habitat suitability for Cusk (Michelle Greenlaw - Fisheries and Oceans Canada) and Cod (Stanley).
- We will also use the stratified abundance estimates from DFO's research trawl surveys for IBR calculations since these data are trusted, available for many species, and standardized across DFO regions.
- The trawl survey data has depth limits (~300m?)
- Ellen Kenchington (Fisheries and Oceans Canada) has more habitat suitability models, mostly for corals. These may reflect historical sampling patterns more than actual coral distribution patterns. We have decided not to use these data for this phase of the case study.
by Remi Daigle
Proposed project timeline:
- Connectivity data transfers completed!
- Draft Marxan Connect analysis completed by Sept 28
- Meeting to discuss draft results scheduled for Sept 28
- Connectivity data transfers completed!
- Draft Marxan Connect analysis completed by Sept 28
- Meeting to discuss draft results scheduled for Sept 28
by Remi Daigle