SCAR/CCAMLR Retrospective Analysis of Antarctic Tracking data (RAATD) - White Paper

A joint project of the SCAR Expert Group on Birds and Marine Mammals, CCAMLR, and the SCAR Expert Group on Antarctic Biodiversity Informatics

draft version, July 2013

Yan Ropert-Coudert, Mark Hindell, Philip Trathan

Scope

The intention of this White Paper is to initiate discussions on the deployment of the RAATD project within the SCAR EG-BAMM community. Please use the embedded comments functionality at the bottom of the page to provide your feedback.

The Retrospective Analysis of Antarctic Tracking data (RAATD) is a joint SCAR CCAMLR project coordinated by the SCAR Expert Group on Birds and Marine Mammals (EG-BAMM). The project aims to undertake a multi-predator assessment of habitat use in the Southern Ocean. Identifying the basic habitat requirements of Antarctic predators is fundamental to understanding how they will respond to the human-induced challenges of commercial fisheries and climate change. This understanding can only be achieved if the underlying linkages to physical processes are related to animal movements. Briefly, this study will develop global and regional habitat usage maps for key species based on tracking data and on the physical and biological attributes of their "hot-spots" and then overlay all the species specific maps to identify multi-species areas of ecological significance (AES). This is a new approach that, by virtue of identifying regions that are important to multiple species, will provide a much better understanding of the regions and processes that require monitoring and management in the future.

The purpose of this white paper is to outline the steps that will be taken to achieve this broad objective, starting from  the identification of likely animal tracking data sets, their collation and archiving, the processing and quality control of the data and the subsequent modeling of AES based on habitat usage maps.

Steps

Step 1:  Data discovery

The RAATD project team will contact potential data holders to compile a metadata catalogue of existing tracking data for Southern Ocean species. The potential data providers will be invited to contribute to the RAATD. Establishment of appropriate data sharing arrangements at this time is a high priority. Data providers can choose to make their data openly public (by being archived on the SCAR Antarctic Biodiversity Database AntaBIF) or share them only with the RAATD project team.

Step 2: Data collation

Eight editors will collate available tracking data and check the quality for key species:

• Elephant seals Mirounga leonina (Ed: Mark Hindell),
• Macaroni Eudyptes chryosolophus and Chinstrap penguins Pygoscelis antarcticus (Eds: Phil Trathan)
• Antarctic fur seals Arctophoca gazella (Ed: Mary-Anne Lea),
• Weddell seals Leptonychotes weddelli (Ed: Dan Costa),
• Emperor penguins Aptenodytes forsteri (Ed: Horst Bornemann),
• Adélie P. adeliae and king penguins A. patagonicus (Ed: Yan Ropert-Coudert),
• Wandering albatrosses Diomedea exulans (Ed: Jose Xavier),
• Humpback whales Megaptera novaeangliae (Ed: Ari Friedlander).

These species were chosen on the basis of the geographic extent of the data and number of animals tracked. The list is not necessarily limited to these species. Species will be added if there are deemed to be sufficient data, geographic coverage or ecological interest.

Step 3: (Meta)data archival on the IPT

To ensure standardised file structure, secure (meta) data storage and facilitation of community access to the data (where appropriate),  the resulting data sets will be uploaded to the Integrated Publishing Tool (IPT), the accepted route for including  data onto Antarctic Biodiversity Information (AntaBif). This has the additional advantage that individual data providers can chose to publish their data sets as a data paper (Chavan & Penev, 2011). Collated (meta)data from STEP1 will be described, standardized (DarwinCore format (DwC) + DwC-Birds and Mammals (DwC-BAMM)) and uploaded using a registered IPT instance. The DwC core terms are listed at the bottom of this document. A provisional set of tracking-specific fields compose the new BAMM extension to the DwC-BAMM format. The proposed fields for DwC-BAMM include the following additions:

Trip_ID	an identifier (typically “trip number”) that distinguishes individual trips during a single deployment event
Hour	the hour in which the position estimate was made
Minute	the minute in which the position estimate was made
Second	the second in which the position estimate was made
ARGOS_Location_Quality	the ARGOS Location Code (3, 2, 1, 0, A, B, Z) associated with the position fix (ARGOS only)
Latitude_Uncertainty	error on the measure of latitude (m) for GLS/GPS
Longitude_Uncertainty	error on the measure of longitude (m) for GLS/GPS
Device_ID	identifier of the tracking device (e.g. serial number)
Device_Type	general type of tracking device used (e.g. GPS, GLS, Argos PTT,...)
Device_Model	the particular device model used (e.g. Wildlife Computers Splash Tag Mk 1, SMRU CTD tag)
Deployment_Latitude	latitude (decimal degrees) at which the tracking device was deployed
Deployment_Longitude	longitude (decimal degrees) at which the tracking device was deployed
Deployment_Locality	locality (place name) at which the tracking device was deployed
Processing_Method	method used to obtain the position estimates from raw data (e.g. “tripEstimation in R”)
Number_of_Satellites	number of satellites used for the position fix, as reported by the GPS tag

Step 4: Raw data publication

Using the IPT, individual data providers that chose to make their data openly public can publish their data sets online via the IPT which will make them available to relevant biodiversity information networks such as ANTABIF, GBIF, OBIS, or BirdLife International. The data will be published as DarwinCore archives.

Step 5: Data preparation

The data from all species will be aggregated. Each track will be divided into individual foraging trips, and then processed to provide the most likely path and its associated uncertainty. Two possible approaches will be to use Kalman filters, or state space models. These cleaned and processed tracks will form the basis of the subsequent modelling.

Step 6: Modelling

A habitat selectivity approach is one possible method for the modelling component of this work. These methods aim to identify the particular environmental conditions that are favoured by the animals, relative to the range of conditions that is available to them. This first requires estimation of the geographic space available to a given animal, which can be done by various methods (e.g. Wakefield et al. 2010, Raymond et al. in prep.). This region of geographic space has an associated range of environmental conditions. Regression modelling can then be used to identify the environmental covariates that discriminate areas that are utilized from those that are not. Species data may need to be subdivided by sex or breeding stage, depending on whether or not the animals respond differently to environmental conditions across such factors. The individual habitat preference models can then be combined to provide a multi-species view on important regions of habitat and their underlying environmental processes.

Step 7: Publication, outreach and uptake by policy organisations

Several publications will result from this work: at least two synthetic papers in scientifc journals and a SCAR/CCAMLR report, perhaps in the form of an atlas similar to that produced for the Patagonian Shelf (http://atlas-marpatagonico.org/the-atlas.html).

EG-BAMM, through its Outreach sub-committee interacting with APECS, will disseminate the outputs of the RAATD project, for example via its “webinars” (online seminars with schools).

The habitat usage maps and the Areas of Ecological Significance will be used by CCAMLR in its spatial conservation planning and other policy organizations, such as the Antarctic Treaty’s Committee on Environmental Protection.

Resourcing:

1. A post-doctoral level researcher will be needed for the habitat modelling and paper writing component of the study. Funding will be sought from ARC and NSF for a 3 year position starting in 2014. There is also the possibility of a Humboldt post-doctoral fellowship.

2. A database specialist will be needed to work in parallel with the ANTABIF team. Funding will be sought through the Belgian Science Policy Office or the Lifewatch e-infrastructure.  

Relevant links

EG-BAMM: http://www.egbamm.scar.org/

EG-ABI: http://www.scar.org/abi

ANTABIF General website: http://www.biodiversity.aq

ANTABIF Integrated Publishing Toolkit: http://ipt.biodiversity.aq

ANTABIF Data Portal: http://data.biodiversity.aq

References

Chavan V, Penev L (2011) The Data Paper: a Mechanism to Incentivize Data Publishing in Biodiversity Science. BMC Bioinformatics 12 (Suppl 15): S2. doi:10.1186/1471-2105-12-S15-S2.

Wakefield ED, Phillips RA, Trathan PN, Arata J, Gales R, Huin N, Robertson G, Waugh SM, Weimerskirch H, Matthiopoulos J (2010) Habitat preference, accessibility, and competition limit the global distribution of breeding Black-browed Albatrosses. Ecological Monographs 81:141-167

DwC core terms

institutionID

datasetID

basisOfRecord

individualID

sex

lifeStage

reproductiveCondition

year

month

day

decimalLatitude

decimalLongitude

scientificName