I always try to make the best of every situation… if you cannot change the circumstances, maximise the use and rescue whatever you can from what is happening. That was my attitude while at home during the covid pandemic, so I participated in all training, workshops and seminars I was able to.
Examples of data types aligned with the five points on the Data Spectrum
One of the most interesting ones, not just because of the topic, but also because of the calibre of the people involved, was a workshop on fisheries data confidentiality, with a few sessions from 30 June to 15 July 2022, organised by the Australian National Centre for Ocean Resources and Security (ANCORS) and Global Fishing Watch.
It was an interesting exercise as it included academics, industry and a couple of consultants. The outcomes became a report that a few participants compiled under Kamal Azmi's lead. The report was presented as a paper to the forthcoming meeting of the WCPFC, and I quote here some of the key parts, yet as usual, I recommend you read the original from here.
Summary for Policy Makers
Transparency is generally seen as good for governance by strengthening government accountability, reducing incentives for corruption, improving the quality of decision making and strengthening trust in governance processes. In extractive sectors, such as fisheries, transparency can help to ensure compliance, build a social licence to operate and build consumer trust for market advantage. Conversely, confidentiality of data and information can erode social trust in institutions.
Greater availability of fisheries data can serve multiple scientific, management and compliance objectives. Data are arguably more powerful when combined with other data by multiple users. Fishery managers and a wide range of stakeholders can improve their understanding of the fisheries and fishery resources, the level of compliance with conservation and management measures and the effectiveness of management regimes. Fishers are, in turn, more likely to trust that compliance is effectively enforced. Third parties, such as researchers, can add to the scientific rigour applied to the fishery and support more effective management. And consumers also increasingly want to know that the seafood they eat has been sustainably and ethically caught and processed.
“Data sharing” does not necessarily mean data are publicly available. Fishers are typically required to provide data to national fisheries authorities and, in regional fisheries, to regional fisheries management organisations (RFMOs). International fisheries law creates duty to share or exchange fisheries data with other relevant States and RFMOs. But data-sharing requirements usually do not mean that data must be shared publicly as open data.
Declaring data as confidential does not explain why. Data are often treated confidentially with good reason. For example, domestic laws typically protect fisheries data under fisheries management and privacy legislation. But in some cases data are regarded as confidential simply because they have been declared to be so. This circular argument conflates “confidentiality” – the extent to which data are protected – with “sensitivity”, which refers to the extent to which harm may be caused if data were to be released more widely.
A recent workshop asked representatives of fishing vessel operators and owners, fish buyers and other stakeholders what makes fisheries data sensitive. The workshop was premised on an assumption that some data may reveal vessel-specific competitive advantages, such as skills and knowledge, proprietary corporate information or intelligence and personal information. The workshop sought to investigate this assumption more deeply. There are many types of data generated and collected in fisheries. In this workshop we focused on a small subset of data related to fishing activity in surface and midwater pelagic fisheries:
Vessel identity and ownership
Vessel location and movement
Vessel authorisations and licences
Transhipments
Catch and effort data
Some dimensions of data could, in some cases, be altered to reduce sensitivity. These dimensions could include spatial or temporal aggregation, specific redactions and alteration of context-specific differences (e.g. gear type or management regime), and could be adjusted individually or in combination.
Vessel identity, corporate ownership, authorisations and catch and effort data are generally not seen as sensitive in isolation. However, sensitivities would arise when ownership was linked to individuals due to privacy and personal security concerns. Vessel authorisations are also not regarded as sensitive. They are readily available on the websites of RFMOs and are trusted sources of updated information about the fishing vessel’s ownership and history. Catch and effort data when combined with vessel identity are highly sensitive as they provide insights into a vessel or fleet’s operations.
Vessel location and movement data are particularly sensitive but can vary with the type of gear or the type of fishery. Sensitivity was largely due to the risk of competitor vessels taking advantage of the skills and experience of other fishers. Sensitivity of fishing location data diminishes over time but opinions vary as to the length of time after which data could be made public. These differences could be explained by the context, including gear types and fishing strategies employed, the type of management regime orfor different target species. For example, pelagic longlining activity data was regarded as less sensitive after a single trip while the sensitivity of purse seine activity diminishes after 60 to 90 days, depending on the operations of the vessel. Spatially aggregated data ranging from 1x1 degree to 5x5 degrees is also considered to be less sensitive while still useful for most research purposes. Location data are much less sensitive in fisheries governed by individual output (i.e. catch quota) controls or where there was a high level of cooperation among fishers.
Transhipment data are both sensitive and a source of frustration for operators. Business intelligence could be gathered about a vessel’s or fleet’s operations by piecing together transhipment data with effort and trip length data. However, a lack of transparency and inconsistent transhipment reporting across fleets means that non-compliant activities were likely going undetected.
Fisheries data were also reported to be sensitive due to operational and business intelligence concerns. Location data is particularly valuable to competing fleets while combinations of catch, effort, transhipment and trip length data and other publicly available information, such as market prices, could provide intelligence about the operations and profitability of a fishing enterprise. That said, some operators are sceptical about the reliability of the assumptions on which such intelligence gathering is based.
Good operators who are more transparent fear they will be more vulnerable to unjustified accusations of bad behaviour. Publicly available data could be misinterpreted, damaging the reputations of good actors and in some cases, leading to security concerns for individuals and assets. Paradoxically, bad operators would carry on unhindered as little information was known about them.
Fishers support full transparency in the long-term but this is tempered by current inconsistencies in the level of transparency and compliance among different operators and fleets. The pathway toward greater transparency needs to ensure that all participants in a fishery are subject to the same reporting and data sharing requirements and that these are enforced fully and consistently. This would mean that a more transparent operator is not placed at a competitive disadvantage compared to a less transparent operator. An important first step would be to ensure that all operators meet current reporting and data sharing requirements to the same standard.
Vessel operators are more likely to respond to calls for greater transparency if the rationale is clear. Further research is required to more fully understand what data are not available to third parties and why they should be available. Identifying ways to handle data without particular fields, such as vessel identity, may also warrant further investigation. And further work could be conducted into whether releasing vessel monitoring system (VMS) data more widely after 60 to 90 days would be acceptable to vessel operators, and under what circumstances.
Conclusion
Increasing calls for transparency make intuitive sense in the quest for more effectively and sustainably managed fisheries. Citizens seek greater accountability in the way public resources are exploited by private interests. But data and information about an individual vessel’s or fishing enterprise’s operations hold considerable commercial value to the enterprise and its public release would represent a loss of value. It may also increase risks or add new risks that were previously low or non-existent. Steps toward transparency must therefore consider commercial and other sensitivities that underlie the confidential treatment of fisheries data, such as vessel identity and ownership, authorisations, vessel location and movement, transhipment and catch and effort data.
This workshop aimed to identify some of the key sources of sensitivity. We identified some clear grounds for sensitivity, particularly in relation to personal privacy and security, and data that reveal skill, knowledge and experience that in turn provide certain operators with a competitive advantage. Sensitivity may vary in fisheries with catch-based controls or a high degree of cooperative behaviour but the exact boundaries require further investigation. The passage of time or spatial aggregation can help to reduce sensitivity but it is less clear exactly what values and under what circumstances this would occur.
Recent developments in the public accessibility of AIS data have meant that the recent and historical location of large fishing vessels is becoming increasingly difficult to hide. Overall, fishers support greater transparency but the pathway to that longer-term objective must not place more transparent operators at a competitive disadvantage. This likely means that the transparency playing field should be levelled by bringing all actors into line with current transparency and reporting requirements before progressing further toward full transparency. Operators also expect third parties who call for greater transparency to make the case for who needs to know, why they need to know and how sensitivities will be managed.