Data & API Requests

Goal: Include FxO as metadata in the Infrastructure API. SkyTruth provided the verified list of which structures are this FxO type for our GFW collaboration in July '24, but the types are not yet available to shared Infrastructure API consumers. Background: For GFW, as a part of our Open Ocean subgrant deliverables, SkyTruth manually developed a verified dataset of a more specific type of offshore oil infrastructure, the category of FxOs. (Deeper sea oil exploration requires particular types of vessel-like, moveable, floating oil platforms, including Floating Storage and Offloading (FSO), Floating Storage Production and Offloading (FPSO), Floating Liquefied Natural Gas (FLNG), and Floating Storage Regasification Unit (FSRU) vessels, which we've previously both referred to collectively as FxOs.) In total, our analysis confirmed 301 FxOs with known operating locations; this information was shared with GFW in July '24. Issue: It is not yet stored as metadata anywhere in GFW's Infrastructure API, which was the planned use case for this open dataset (so that our shared impact partners on this topic can all access the new data.) SkyTruth itself relies on GFW's Infrastructure API and Vessel API to display sources of pollution in the Cerulean app. As FxOs are of interest and the topic of our forthcoming co-authored GFW-SkyTruth white paper, we'd like to be able to display the most accurate infrastructure type, while using the GFW API as we currently do. Note on related issues: this request is categorized as vessel identity because of the structure ID/vessel ID overlap in GFW APIs today that is common with FxOs (given they are former vessels being used as infrastructure) -- Please see other tickets about duplicates between APIs. Thanks! -Kris M.

At SkyTruth, we identify sources of marine pollution using both vessel and infrastructure data from GFW (both can cause pollution). When a stationary facility - such as an FPSO or an oil platform - is broadcasting AIS, it will appears on both GFW APIs, resulting as duplicate potential sources of pollution. This is misleading for advocates and journalists investigating the cause of the detected pollution, as it looks like there are 2 possible polluters when actually there is 1 likely culprit. We would prefer to use the AIS version of the two because includes additional metadata about the AIS broadcasting facility (flag, vessel name, etc), however it is generally more accurate to describe these locations as infrastructure and not vessels, so it would be nice to append the Structure ID somehow so we can describe it correctly. (Oil slicks that come from stationary sources also look quite different from oil slicks cause by vessels in motion, so this is also useful for pollution investigators.) Is it possible to cross-reference the lat/lon of known infrastructure with known AIS to de-dupe somehow? Examples (not exhaustive, we see this relatively frequently, especially among FPSOs/FSOs and similar floating infra) -- We also have the S1 scenes for these locations available if helpful for your investigation. Structure ID 239392 & MMSI 309999000 (also shared in screenshot) https://cerulean.skytruth.org/?zoom=13.578009&lng=11.269523&lat=-6.146767&startDate=2024-06-03&endDate=2024-12-03&cls=2_3_4_5_6_7_8&sources=1_2&sourceScore=-1_&sourceLimit=3&imageGamma=2&imageOpacity=1&machine_confidence=0.9&area=5_50&mpa=0&eez=0&iho=0&hitl_cls=&offset=0&limit=100&hitl_rev=0_1&slickId=3573116 Structure IDs 340743 & 88494; MMSI 2320206. https://cerulean.skytruth.org/?zoom=11.435713&lng=1.471807&lat=58.078448&startDate=2024-06-03&endDate=2024-12-03&cls=2_3_4_5_6_7_8&sources=1_2&sourceScore=_&sourceLimit=5&imageGamma=2&imageOpacity=1&machine_confidence=0.9&area=5_50&mpa=0&eez=0&iho=0&hitl_cls=&offset=0&limit=100&hitl_rev=0_1&slickId=3573183 Structure IDs 1040285 & 77739; MMSI 235599000. https://cerulean.skytruth.org/?zoom=11.425581&lng=0.914865&lat=57.087792&startDate=2024-06-03&endDate=2024-12-03&cls=2_3_4_5_6_7_8&sources=1_2&sourceScore=_&sourceLimit=5&imageGamma=2&imageOpacity=1&machine_confidence=0.9&area=5_50&mpa=0&eez=0&iho=0&hitl_cls=&offset=0&limit=100&hitl_rev=0_1&slickId=3573376 Structure IDs 592009 & 457875; MMSI 538003299. https://cerulean.skytruth.org/?zoom=12.915541&lat=28.466801&lng=49.367199&startDate=2000-06-04&endDate=2024-12-04&cls=2_3_4_5_6_7_8&sources=1_2&sourceScore=_&sourceLimit=3&imageGamma=2&imageOpacity=1&machine_confidence=0.5&area=0_&mpa=0&eez=0&iho=0&hitl_cls=&offset=0&limit=100&hitl_rev=0_1&slickId=3558829 --Kris

At SkyTruth, we've observed duplicates in the Infrastructure data where single pieces of infrastructure appear multiple times in the DB. When comparing with Sentinel-1 imagery, only a single piece of infrastructure appears to be present. To see in Cerulean (which relies on the GFW API), view the infrastructure locations on the map with the squiggly line button next to each infrastructure source: https://cerulean.skytruth.org/?zoom=11.419613&lat=57.192989&lng=0.99397&slickId=3573395 - Also shown in screenshot. Structure IDs are 352531 and 116096 in this example, but we observe this relatively frequently. This is an issue for our advocates and journalists investigating causes of marine pollution because it looks like there are 2 possible polluters when there is actually just 1 likely culprit. --Kris

From coupdemistral@gmail.com (Mauro Calvi): "Have you considered making bunker-to-bunker encounters available in addition to the combinations of vessel types you currently support? These transshipment operations are often illegal and represent serious environmental threats. I appreciate this may be outside of the scope of your mission, but it is in line with the broader mission of using your data to protect our oceans"

Feedback: • Fishing Effort Data Access and Analysis: I was able to use the API again to look at encounters of boats and see if our birds were associated with boats that have fishing effort data. It’s been eye-opening. The ability to see a vessel track in tandem with an albatross track allows for so many questions to be answered from a perspective that wasn’t available before. • Real-Time Data Visualization: I think it’s worth mentioning that while the portal has a delay in real-time data (around five days), it’s still incredibly useful for rolling through time series and visualizing interactions. The improvements to the map, along with the API, have been really helpful. • GFW Map and API Improvements: I really like the brighter colors, as the previous lighter colors were a bit of a strain on the eyes. I was happy to see I could upload bird tracks as points rather than tracks, which was especially useful for visualizing radar pings. I made the radar points larger to indicate when a bird was near a boat, and that was really cool to see in real-time. Lastly, I’d say that without the API and portal, it would be impossible for us to visualize and interact with this level of vessel and bird data. Global Fishing Watch has built a tool we couldn’t replicate on our own, and it’s been invaluable for our research • Educational Use Cases: I’ve also been using the portal to set up another professor with albatross data for a lecture and lab format at San Jose State University. Product Requests and Other: • Inclusion of Non-Fishing Vessel Data: One thing I’d like to see in the API is access to vessel density, not just fishing effort. I’d be interested in something similar to the density layer for container ship traffic or general vessel density, not limited to actively fishing boats. It would be helpful to have access to composite densities for all AIS data, in line with what your portal offers. • Handling International Date Line Crossings: One thing I’ve noticed is that when birds cross the International Date Line, it causes issues if the tracks are uploaded as lines, but uploading them as points works fine. For a recent assignment, I provided my students with two example tracks with clear radar hits associated with vessels and four tracks with radar hits that didn’t have an associated vessel. Their task was to use the portal to investigate whether those hits were truly unmatched or if they could identify vessels with gaps in AIS transmissions.

FFA and others in the Pacific mentioned they'd like to check it out to possibly help with their patrols, with a particular interest in RFMO authorizations. More context: They're specifically looking to be able to view recent RFMO authorizations for vessels that enter into their RFMO's waters. I think the API or gfwR wouldn't be the best fit for their skills/time availability because they seemed more interested in a basic excel-sheet-style database that an entry level staff member could quickly search for a vessel's recent authorizations. Maybe the functionality they're looking for is already available in a product such as vessel viewer, but they seemed keen to access the underlying data a bit more than that.

I am wondering if it is possible to access the "Environment" data layer in addition to the vessels and fishing activity layers via the APIs. I was hoping to use your api in conjunction with the vessels and fishing activity layers to augment a machine learning project for microplastic concentration prediction

Our lab has been looking into trying to use the GFW data products for some of our research projects. We have a number of recorders in the ocean, and we would like to be able to use AIS data to try and help quantify the impact of vessels on the soundscape. The 4Wings API has been awesome to try and get at fishing activity, but I think for areas where fishing vessels are not the main source of vessel noise this is probably not telling the full story. Do you know if there are any plans to make the pink "Vessel presence" from your map tool available to the 4Wings API? Or is there a way to get that data that I'm missing? I saw the note on your website about not being able to make the raw AIS data available, but I was hoping it might be possible to get a raster summarised version with presence hours instead of raw AIS locations.

I am doing some research about predicting vessel trajectory. I try to get [mmsi, lat, lon, sog, cog] of multi vessels from https://globalfishingwatch.org/map , but I only can download a vessel's data each time. So how do I can get all of those data of multi vessels? Could you help me in this situation? I look forward to hearing from you! https://github.com/GlobalFishingWatch/gfwr/issues/171

● Vessel Identifiers: Vessel events are tied to a vessel ID that is assigned by GFW. When a different set of event data is incorporated into GFW’s database, the same vessel may be given a new GFW vessel ID. FishWise suggests having a vessel identifier that does not change and could be followed, especially in the cases of vessel name changes, flag changes, etc. ● API with ISSF: To understand which vessels are engaging in other sustainability initiatives and/or undergoing additional assurances, provides important context to the Vessel Viewer insights and IUU indicators. FishWise recommends GFW establish an API connection with ISSF’s PVR and VOSI lists to, at minimum, confirm the presence or absence of vessels on these registries. ● Sunset/Sunrise Time: Environmental sustainability is a key consideration, especially potential risks of interactions with endangered, threatened, and protected (ETP) species such as sharks, sea turtles, marine mammals, and sea birds in tuna fisheries. Adding daily sunset and sunrise times to the GFW database can help assist with deducing the potential risks of fishing activities interacting with ETP species.