NFFM73102301304KAW Reponses to Questions.pdf

NFFM73102301304KAW

RESPONSES TO QUESTIONS

Start-up project scoping, data collection, data analysis, algorithm development, software production, process/procedural documentation and report writing to support the Northeast Clam Survey Program

NOTE: This list is a compilation of all questions received pertaining to the contract opportunity listed above. The newest question(s) and response(s) is in blue text.

1. Does the government have interest for an alternative commercial off the shelf (COTS) solution?

Yes, any robust solution as long as NOAA does not have to pay subscription or other fees would be considered.

2. How many different vessels (sampling stations) will be collecting data simultaneously? I’m asking in order to estimate how many total cameras & equipment would be needed. Sounds like it would be much more than 12 (12 strata) since page 2, Section C.2.1. says "The survey is conducted by randomly assigning survey stations to individual clam strata (Figure 1)..."

There is only one vessel. Stations refer to sampling events (i.e. a location selected for sampling). The vessel generally visits about 180 stations during a survey.

3. Would you happen to have any information on time length/frequency estimates?

This would help me to estimate the Cloud Storage size & costs for video storage.

Example of info that would be helpful:

a. Average length of time that a full sampling session takes, per leg

b. Total number of legs across the duration of this project

The average amount of video time needed per station is on the order of 20 minutes. This is the time that clams are running down the conveyor belt.

There are about 60 stations per each of the 3 five day legs. The cameras do not need to run when there are not clams on the belt. So total video time could be about 2 hours per day for 15 days or a total of 30 hours.

This is an approximation and having a healthy buffer would be prudent.

4. Do you happen to have any historical information on volume approximations for the amount of clams harvested during a dredge survey [either averages or ranges]? This will help me estimate how many images I might expect to have for training.

This is a table of summary stats for the total number of clams of each species caught at each tow in one year in the same area that will be surveyed this summer. So in brief an average of 115 surfclams and more than 4000 quahogs were caught at each of the ~180 stations. There should be plenty of images.

Min. 1st Qu. Median Mean 3rd Qu. Max.

Surfclam 0 0 3 115.5 40 2500 Quahog 0 86 1218 4482 5530 58053

5. Does surveying occur for one month only? Just the month of August only and not beyond August?

Correct.

6. Does clam surveying happen in the daytime only or at nighttime as well? What time ranges of the day are survey vessels deployed?

Surveys are conducted 24 hours a day, continuously during each 5 day leg.

7. Does the software have to automatically detect when to begin recording, or do skippers (or boat personnel) manually trigger the camera to start recording?

Either option is possible. We would prefer automatic start stop (it could be tied to the conveyor belt being turned on), but we have a lot of other equipment that gets turned on for each station so manual start stop is a possibility - although I would expect this to increase the needed storage space due to human error (forgetting to turn the camera off seems a strong possibility).

8. Does the port have wifi available? If yes, any technical specifications regarding internet speed (upload / download speed) would be helpful as well. I'm asking to understand more about transmission options after docking.

The boat will dock in New Bedford between legs. The exact location is not known ahead of time so while wifi is a possibility, it is not a guarantee.

Alternatives should be planned for (e.g. downloading to hard media).

9. What is the make and model or brand name of the vessels that will be used in surveying?

The survey vessel is the ESS Pursuit (there is some information available through an Internet search using the vessel name) but it is a purpose-built clam fishing vessel (brand name unknown). The vessel’s length overall (LOA) is 48 meters and width is 15 meters. There is a photo of the deck and conveyor in the Statement of Work (SOW).

10. Section ‘C.3.2 Algorithm Development’ mentions (twice) that we’re expected to leverage existing datasets. It says 1) “The machine learning components are expected to utilize the existing image data sets” and 2) “The goal is to leverage existing data sets as much as possible for algorithm development, while also highlighting future image training set needs.” Is section C.3.2 alluding to there being another pre-existing clam image dataset (not fish)? I ask because based on Section C.3.1, I understand that the Contractor is supposed to develop the clam image dataset.

The images will be generated by the contractor.

11. Where can I read about the ‘Options’ that are mentioned in Section (IX)(b) of the

‘Combined Synopsis’ doc. I didn't notice any Option items mentioned in the

SOW.

Correct. There are no Options so that language is not applicable to this

SOW.

12. During the image gathering phase, for the Image Library, will there be a human auditor on board? I ask because I’d like to know if a human's findings will be provided to the Contractor to enable proper model testing/validation of the model that I develop.

Yes NOAA expects to work closely with the contractor to provide species identification and enough auditing of images to make this project feasible.

There will also be many stations where the normal survey data collected will be useful. For example, when the tows are small enough, all of the clams are identified, counted and measured. That information can be made available for training the algorithm.

13. Section C.4.3. says that “The Contractor shall also have adequate spare or backup equipment available locally in case of a system failure or equipment malfunction”. Does ‘locally’ mean that backups should be aboard each vessel?

Or does ‘locally’ mean that backups should be available at the port?

There is only one vessel, so onboard the vessel would be the preferred option. If there is a failure that cannot be corrected by the staff on the vessel the contractor would need to meet the vessel in port between legs to make repairs. Communication with the boat will be limited during the survey (satellite phone only and intermittently) so leaving instructions for common repairs will be important.

14. Section C.4.4. mentions the ‘ESS Pursuit’. Is this the vessel that is expected to be used across all surveys?

Yes. This is the only boat and there is only one survey.

15. Section C.4.4. says “The Contractor shall develop a clear protocol for image and specimen data collection and pairing. The protocol shall outline suggested specimen handling protocols and camera equipment configurations being used throughout the project. The study protocol document is a communication tool used to ensure that NMFS staff, Contractors, field technicians, and project coordination staff know what their roles are for a successful project implementation. The protocol document will include at a minimum; general image collection procedures, camera system configuration, operator responsibilities, protocols for data retrieval, specimen handling protocols, general equipment information, basic camera and data system troubleshooting, project objectives and contacts, and any other pertinent information.” Please confirm if my assumption is correct. Is this protocol to be created after the contract is awarded? And not to be submitted along with the quote?

Correct. The protocol will be created with some collaboration from Northeast Fisheries Science Center (NEFSC).

16. Can I invoice every month? As opposed to just 1 lump sum payment after 12 months? As a small business operation this would help tremendously to pay the team.

FAR 52.232-1 will be included in the purchase order and it addresses payments.

17. Could you please provide the excel sheet template for the ‘Schedule of Items and Prices’ as a separate attachment?

Schedule of Items and Prices is not mentioned in the Combined Synopsis or the SOW. If the question is asking for clarification for how to provide the quote price, itemization of items and prices is not required nor is there a specified format for presenting the quote price.

18. Could you please provide historic, step-by-step images of the clams being processed? I’m asking because I’m trying to visually understand what clam processing in action would look like according to the SOW: “Positioning of the clams on the conveyor may pose a challenge if the catch is large as clams may pile up and may orient vertically…”.

Please refer to the document ‘Processing Clams’ that has been added to this posting.

19. How tall is the camera mounting structure above the conveyor belt system?

There is no camera mounting structure. There are currently handrails and short walls (see photo on the above webpage for hand rails and inserted images for side walls). This is however a functioning commercial fishing vessel and the crew can weld anything we want onto the conveyor structure. The conveyor also serves as a walk way, so ideally, the camera will sit either very low over the belt with a protective cover over it, or above head height. The low option is preferred because the camera would then only see the belt and the crew probably does not want to be in the images.

20. Does the ship have internet connectivity during the survey?

No

21. Is there already a light that is used over the conveyor belt? If so, what are the specs?

The camera will need to have its own light source.

22. If there is a light above the conveyor belt, what is the height of the light from the conveyor belt?

See the response to question 21.

23. Where is the access to a plug for power either near or above the conveyor belt?

An extension cord can be ran to the camera if needed.