• Sophie Pierszalowski

    Sophie Pierszalowski
    Univ. of Washington
    Tracked humpbacks and other whales around Kodiak Island in Alaska.

    This summer, while working in the National Marine Mammal Laboratory at the Alaska Fisheries Science Center, I was able to take part in an amazing field study opportunity around Kodiak Island, Alaska.

    There were three components of scientific research that occurred during the field portion of this internship. The Kodiak Island whale cruise took place from July 7th- July 18th, and mainly consisted of work done in Uganik Bay, Uyak Bay, and Izhut Bay. During this time we collected biopsy samples from various fin and humpback whales, we tagged one fin and five humpback whales with satellite transmitters, and we collected photo identification data for every whale we encountered throughout the trip in order to identify each individual. Biopsy information will be used for future analyses while satellite tags have provided, and continue to provide, detailed information on location and movement of the whales.

    My project was to analyze and curate the photo-identification data and to assign each individual its own identification code. Matching individual whales using digital photographs allows us to analyze movement, behavior, kinship, survival, and other life history traits. My portion of this project, completed during both the field and office components of this internship, involves streamlined management of biopsy, photo ID, and satellite tag data collected throughout the trip. My contribution to the larger scale scientific analysis of humpback whales around Kodiak Island is based on organizing 4,400 digital photographs into a comprehensive database that can be queried by searching information

  • humpback whale

    The humpback whale we nicknamed "Strokes" was tagged in Izhut Bay on July 15th and was still giving us signals over four weeks later. Photo by Sally Mizroch, NMFS Permit No. 782-1719

    embedded or entered into the EXIF metadata in each photo image file; e.g., GPS location, date and time the photo was taken, the whale's field nickname (e.g., "Fuzz" & "Strokes"), the encounter number, the part of the whale's body that is shown, the biopsy sample number, and the credited photographer.

    There are many steps involved in the management of this extensive dataset. Not only are there 4,400 photos, but each photo has seven or eight metadata components. Metadata categories require manual assignments of credited photographer, encounter number, body part, precise location, and individual nickname. These are all deciphered through analysis of each individual photograph.

    After allocating the proper information to the appropriate categories, the photos are edited in ACDSee Pro 2 in order to maximize the readability of each image. The images are then converted from RAW files to jpgs at which point they will maintain the quality of the original RAW file but will be more compact and therefore will display more quickly when viewed using hyperlinks in the database.

    Once the field data have been organized into a relational database, the data can be analyzed using simple database queries.

  • map

    For example, even with this relatively small dataset, I will be able to analyze:


    • frequency of sightings of each individual whale

    • the duration of each whale encounter

    • group information such as group size, success rate of how many whales in a group were identified, associations of individuals

    • locations each time each whale was photographed. With this information we can compare sighting locations from this project to past sighting locations and the data will be easily retrieved from the database for any future sightings of these whales.


    These data will be integrated into the large photo-identification collection/database developed and managed by my intern supervisor Sally Mizroch, which contains humpback whale photos dating back to 1966.