Better Call Trawl

Dive straight into the data here.

We all like beautiful sandy beaches, and we all love the sea turtles that nest on them, right? But considering climate change and future sea level rise projections, how do we continue to maintain healthy beaches in harmony with sea turtles? - that’s the crux of this challenge. 

To preserve our beautiful sandy beaches, the Bureau of Ocean Energy Management (BOEM) authorizes giant dredges to vacuum up sand from the bottom of the ocean to replenish eroded beaches. In the process of collecting sand, however, sea turtles resting near or on the seafloor can be sucked up by the dredge and can be injured, often lethally. 

Over twenty years ago, a team of experts developed sea turtle relocation trawling methods to try and prevent sea turtles from being injured or killed during dredging. Essentially a fishing net is dragged along the seafloor to capture sea turtles near where dredging is taking place. These captured sea turtles are then moved out of the dredge area to keep them safe. 

The trawling, catch, and relocation of sea turtles has drawbacks. It is thought that only the sea turtles near or on the seafloor are at risk of being sucked into the dredge. But trawling could be capturing sea turtles swimming above the sea floor, as well as any other creature on or near the sea floor, that may otherwise not be impacted by the dredge. Furthermore, the act of catching sea turtles and moving them could injure the sea turtles and is likely stressful. So the trick is to catch and relocate as many at-risk sea turtles (those less than two meters above the sea floor) as possible, without disturbing the non-at-risk sea turtles (those more than two meters above the seafloor) or other creatures that should not be caught.
 

Where you come in:

The efficacy of sea turtle relocation trawling has yet to be quantitatively tested. BUT BOEM is ‘swimming’ in great data - and they need your help! Your use of these unique datasets, and potentially data from other sources, will help BOEM understand more about:

• How many at-risk sea turtles sea turtle relocation trawling actually catches and moves 
• How many non-at-risk sea turtles are captured and moved
• How the efficacy (at-risk sea turtles/ non-at-risk sea turtles captured) changes over the period of dredging; Or based on the behavior of the sea turtles (directed swimming, migrating, resting), etc.
• What additional data could be collected to bolster your confidence in the conclusions
• How current sea turtle relocation trawling methods can be modified to increase at-risk sea turtle catch rate while decreasing the catch rate of non-at-risk sea turtles and other bycatch

The Challenge seeks analytic tools (e.g., decision dashboards, data markdown files/notebooks, analytic reports) to demonstrate projected sea turtle relocation trawling effectiveness, help guide new scientific studies, and improve future sea turtle relocation trawling effectiveness. Up to 4 winners will share in a $40,000 prize purse.

Tackle this challenge and the next time you are enjoying your favorite drink on a white sandy beach you can know that you helped keep beaches sandy AND sea turtles safe!!

The data and other helpful resources are available here.

Challenge Breakthrough

Submit analytic tools and visualizations with data findings and recommend future trawling and data activities.

Background

Sea turtle relocation trawling is used as a tool to move turtles and prevent them from being injured or killed during dredging. However empirical evidence supporting sea turtle relocation trawling as a mitigation tool is limited. During dredging, sea turtles can be accidentally sucked into the dredges, so suction heads are kept on the sea floor while pumps are creating suction. In theory this should limit the accidental capture of sea turtles by dredges. However, anecdotal evidence suggests that sea turtles directly on or 1-2 meters above the sea floor could still be at-risk of being injured, possibly lethally, depending on environmental factors and related operational characteristics of the dredge. To prevent these ‘at-risk’ turtles from being captured by the dredge, a trawling net is used to capture sea turtles and move them out of the way of the dredge. Trawling nets, which are dragged along the seafloor capture sea turtles up to 15ft from the seafloor, as well as any other animals big enough to be caught in the net. This bycatch (i.e., animals that aren’t meant to be captured and moved) such as non at-risk sea turtles (those greater than 2 meters above the sea floor), sturgeon or other animals, can be injured or stressed by capture in the trawl. BOEM’s Marine Minerals Program seeks to quantify the effectiveness of sea turtle relocation trawling (i.e., the ratio of at-risk sea turtles vs. non at-risk sea turtles which are moved), as well as identify data or methodological advancements to improve sea turtle relocation trawling effectiveness.

What is Sea Turtle Relocation Trawling?

Sea turtles are intentionally captured and relocated out of the area of dredging operations to reduce risk of sea turtles being injured by a dredge. Modified shrimp trawling equipment is used to sweep the sea floor to capture and relocate sea turtles. This involves the dragging of a net (dim 15 x 35ft) along the sea floor behind a trawling boat to capture turtles. Typically a vessel with two trawling nets (example image below) is operated in a given area at the same time, and each trawl can be pulled for no more than 40 minutes to prevent serious injuries to captured turtles. Relocation trawling must maintain a safe distance from the dredge and other vessel traffic in the area. Therefore, the trawler is often not working directly in front of the hopper dredge but is instead continuously working to remove species from the general dredging area.  Relocation trawling vessels are smaller than dredges and therefore more restricted by the weather conditions in which they can safely operate. Therefore there are times that the dredge operates, but trawling does not.

Photo credit: East Coast Observers, Inc.

--> Watch now: "How effective is sea turtle relocation trawling" or view the full webinar here.

What is dredging? 

Trailing Suction Hopper Dredges are specialized ships that have a hollow hull which can be filled with sand. Fitted with powerful pumps, the dredges suck sand from the surface of the seafloor through long intake pipes, called dragarms, and store it in compartments in the hull (a.k.a. hoppers). Trailing suction hopper dredges typically have two dragarms, one on each side of the vessel. A dragarm is a pipe suspended over the side of the vessel with a suction opening called a draghead which sits on the bottom and sucks up sand like the head of a vacuum.  Depending on the hopper dredge, a slurry of water and sand is generated from the plowing of the draghead “teeth,” by using high pressure water jets, and the suction pumps. The water and sand slurry is pumped into, and distributed within, the vessel’s hopper where the sand settles to the bottom and the remaining water is pumped back into the ocean. When the hopper attains a full load, dredging stops and the ship travels to either an in‐water placement site, or hooks up to an in‐water pipeline, where the dredged sand is offloaded. The average speed for hopper dredges while dredging is between 1‐3 knots, with most dredges never exceeding 4 knots. The draghead width is approximately 10-16 feet and approximately 6 inches to 2 feet of sand is removed during each pass of the draghead.

With the solutions from this challenge, modeled trawling scenarios using new or varied relocation or turtle avoidance approaches, which have data-backed confidence, will allow BOEM to intelligently test the approaches in the field and make new recommendations on gear or operation modifications that trawlers can make to increase probability of higher at-risk turtle catch rates and lower bycatch. BOEM will better understand the likely effectiveness of a planned trawling operation under different environmental conditions relative to the risk of lethal capture by dredges. Sea turtle relocation trawling operations will be safer and more effective, and better data will be collected.

What You Can Do To Cause A Breakthrough

Build an analytic tool (online dashboard, data notebook or markdown file, statistical model) that leverages BOEM data  including individual sea turtle movement behavior, trawler sea turtle catch data, trawler specification data, environmental and oceanographic data and other resources you find useful, to help BOEM analysts decide how effective current sea turtle relocation trawling methods are and what could be improved by projecting effectiveness.
 

--> Watch now: "What can you do to cause a breakthrough?" or view the full webinar here.

The Data

BOEM has provided two primary data resources (available in the Resources tab) that include:

1. Tag data of individual sea turtle behavior collected using Inertial Measurement Unit (IMU) tags, which have accelerometer and depth sensors and thermometers, attached to sea turtles. There are data from two tagging locations: one in the Gulf of Mexico which has data from three tagged turtles displaying different types of behaviors (e.g., swimming, resting, diving) in shallow and deeper coastal areas, and one in the South Atlantic (Cape Canaveral, FL) which has 8 turtles in a mixture of shallow shoal and coastal areas. Sea turtles were not captured and released in active dredging areas, so it is unlikely that they were captured in a trawl. All behaviors are assumed to be ‘normal’ behavior. Because these individual turtle behavior data and the below trawl data are not taken from the same area, the individual behavior data are assumed to be representative. Additional information about the data and the scenarios of turtle behavior is provided in the Assumption section below.
   1. Gulf of Mexico data are organized into four folders, one for each species and behavior type. Each folder has two .csv files, one containing accelerometer data, and one containing depth and temperature data. A full data dictionary is contained within the data files provided.
   2. Cape Canaveral data contains both accelerometer/depth/temperature data, as well as lat/lon/date/time data, since these turtles were tagged with both satellite location tags and accelerometer/depth/temp tags. These data are organized into eight .csv files. Each file contains time matched GPS and behavior data (accelerometer/depth/temp). A full data dictionary is contained within the data files provided.
2. Trawling data. These data include information on the catch for each trawl for two dredge areas (BA45 and BA143), as well as spatial files describing the dredge area (before and after dredging), and the capture and release locations of sea turtles. Variables include trawl duration, number, size and species of turtles captured.

Solvers can also explore secondary datasets, but should focus on the two primary data sets for exploration. Additional data can include environmental covariates (e.g., oceanographic data (provided here), and dredge data (provided in the challenge data in the Resources tab). Please note that BOEM cannot guarantee the quality and reliability of secondary data as they relate to the primary data.

ACCESS THE DATA HERE.

--> Watch now:  Data Walkthrough

Driving Questions and Assumptions to Validate

In the exploration of the data, solvers should consider the following questions, based on the assumptions listed below, when you submit your solutions:

Questions

1. What could be changed about the trawler equipment and its operation to catch more at-risk sea turtles and less bycatch?
2. What data could be collected in future BOEM field projects to help the analysis and provide even more insight?

Assumptions

Trawling data sets

• Trawling - Sea turtle relocation trawlers typically operate at about 2-3 knots with two nets in the water and on seafloor.  Each individual tow is limited to 42 minutes to limit the risk of adverse effects.  Each net span in the water is approximately 35’ wide (70’ total sweep width) with a maximum  height of approximately 15’ off the bottom at the center of each net tapering to the wings.  
• Sea turtles - Trailing Suction Hopper Dredges are required to keep dragheads on the bottom at all times when the pumps are engaged to constrain the suction field and mitigate entrainment risk. Therefore, during normal operating conditions, only sea turtles on or near the seafloor are at risk of encountering the Trailing Suction Hopper Dredge draghead (within <2 m). US-flagged Trailing Suction Hopper dredges typically operate both port and starboard dragheads.  Draghead dimensions vary depending on the specific vessel design and operating conditions; however, for the purpose of this challenge, assume the draghead width is approximately 10’-16’ for large Trailing Suction Hopper Dredges operating offshore and the cut depth for each pass of the draghead is approximately 6” to 2’ depending on sediment type. Turtle data is partitioned by various behavior scenarios:
  • Swimming
  • Resting
  • Swimming or resting in shallow habitats
  • Swimming or resting in deep habitats
• The site - The dredging site, where trawling occurs, is assumed to be two square kilometers with depths ranging from 28-46 meters. There are anticipated to be a total of  0 - 50 sea turtles in the defined site area at any given time.
• Bycatch - While at-risk turtles are the primary focus of this challenge, BOEM is interested in reducing the risk of catching non-at risk sea turtles, other protected species (e.g., Atlantic sturgeon, small tooth sawfish, and giant manta rays), and non-protected species of fish, sharks, rays, etc.

Tagging data sets

• Triaxial accelerometer tags - measure acceleration in three axes. This data can tell you if the sea turtle is resting, or swimming. X axis = flipper beats, Y axis = dive angle, z axis = roll
• Depth data - tagged sea turtles used both shallow and deep habitat. In general for the Gulf of Mexico dataset you can assume a shallow depth of 2 - 6m and a deepwater depth of 20m. For the Cape Canaveral dataset you can assume a depth of 5 - 20m. Using depth and accelerometer values you can confirm whether the sea turtle is resting on the seafloor, thus telling you the exact depth of the seafloor. 
• Species - while there are 5 species of sea turtles that could be present in the dredge area, there are likely not enough species-specific behavior data to differentiate by species or life stage (juvenile/adult)..
• Location - sea turtles were tagged in different locations but sea turtle behavior is assumed to be representative of sea turtle behavior near the dredging site. 

--> Watch now: "What is BOEM trying to understand?" or view the full webinar here.

The Data Transformation

Solvers will run the data through their preferred and ideal analytic tools to create outputs that help BOEM understand trawling effectiveness and plan future field activities. In the process, explain the labor that goes into preparing data for such analyses. What could be done to the data beforehand to make that transformation work less intensive? What data do you feel, if you had easy access to, would make the process easier and also make the analysis more valuable?

PRIZE

Up to four prizes will be awarded to the best submissions received that provide insight into BOEM data, insight into trawling practices, and inform future data and field activities. The first three prizes are tiered and the runner up prize is for a solution that demonstrates a particularly innovative recommendation for improving methods of trawling for sea turtles.
 

TIMELINE

Open to Submissions:  September 13, 2022

Submission Deadline:  December 15, 2022 @ 5PM ET 

Judging:  December 15  to February 15, 2023

Winners Announced: February 21, 2023

HOW DO I WIN

To be eligible for an award, your proposal must, at minimum:

• Satisfy the Judging Scorecard requirements
• Thoughtfully address the Submission Form questions
• Be scored higher than your competitors!

Solutions should:

• Analyze BOEM data and demonstrate useful insight that can lead to better uses of the data, better trawling method possibilities, or both
• Be analytic tools that BOEM can use to inform their next phase of work that will follow this challenge, ideally testing new methods of relocation trawling based on evidence-based recommendations from this challenge
• Provide specific recommendations on BOEMs data collection and use of data, or trawling methods and ways to improve the relocation practice of at-risk sea turtles, and reduce the risk of bycatch of non-target species (including non-at-risk turtles)
• Use visualization to communicate the analyses and recommendations clearly
• Be scalable and reusable with additional data and for future BOEM projects

Be sure to consider the judging criteria weights in constructing your submission.

JUDGING CRITERIA

RULES

Participation Eligibility:

The Prize is open to anyone age 18 or older participating as an individual or as a team. Individual competitors and teams may originate from any country, as long as United States federal sanctions do not prohibit participation (see: https://www.treasury.gov/resource-center/sanctions/Programs/Pages/Programs.aspx). If you are a NASA employee, a Government contractor, or employed by a Government Contractor, your participation in this challenge may be restricted.

Submissions must originate from either the U.S. or a designated country (see definition of designated country at https://www.acquisition.gov/far/part-25#FAR_25_003), OR have been substantially transformed in the US or designated country prior to prototype delivery pursuant to FAR 25.403(c). 

Submissions must be made in English. All challenge-related communication will be in English.

You are required to ensure that all releases or transfers of technical data to non-US persons comply with International Traffic in Arms Regulation (ITAR), 22 C.F.R. §§ 120.1 to 130.17.

No specific qualifications or expertise in the fields of ecology, dredging, or trawling is required. BOEM encourages outside individuals and non-expert teams to compete and propose new solutions.

To be eligible to compete, you must comply with all the terms of the challenge as defined in the Challenge-Specific Agreement.

Intellectual Property

Innovators who are awarded a prize for their submission must agree to grant the United States Government  a royalty-free, non-exclusive, irrevocable, worldwide license in all Intellectual Property demonstrated by the winning/awarded submissions. See the Challenge-Specific Agreement for complete details.

You may be required to complete an additional form to document this license if you are selected as a winner.

Registration and Submissions:

Submissions must be made online (only), via upload to the HeroX.com website, on or before December 15, 2022 at 5pm ET. No late submissions will be accepted.

Selection of Winners:

Based on the winning criteria, prizes will be awarded per the weighted Judging Criteria section above.

Judging Panel:

The determination of the winners will be made by HeroX based on evaluation by relevant BOEM specialists.

Additional Information

• By participating in the challenge, each competitor agrees to submit only their original idea. Any indication of "copying" amongst competitors is grounds for disqualification.
• All applications will go through a process of due diligence; any application found to be misrepresentative, plagiarized, or sharing an idea that is not their own will be automatically disqualified.
• All ineligible applicants will be automatically removed from the competition with no recourse or reimbursement.
• No purchase or payment of any kind is necessary to enter or win the competition.
• Void wherever restricted or prohibited by law.