Reservations and limitations of the mapping tool “The seafloor footprint of Danish fishing”

The interactive mapping tool “The seafloor footprint of Danish fishing (FPLADS)” displays three main types of results for the area, fisheries, and time period in question: 1) a map of fishing distribution and intensity, 2) table(s) presenting numerical indicators of the fishing distribution and intensity, and 3) a graph showing both annual and multi‑year percentage values of area impact for habitat types and areas of interest.

We describe the most important reservations and limitations of the data below. 

Fisheries data

The mapping tool is based on positional data from VMS, black box, and AIS, which are linked with logbook data from fishing vessels. Using knowledge of typical vessel speeds during fishing with different gear types, the positions where fishing activity is assumed to occur, and thus where trawl gear is in contact with the seabed, are identified.

Information on vessel length and engine power is retrieved from the vessel registry and used to estimate gear width (eventually refer to this scientific article). The estimated fishing positions are combined into vessel tracks, and a buffer corresponding to the gear width is added to delineate the impacted area.

The method generally provides a reliable representation of actual fishing activity, but errors do occur. Since fishing activity is estimated based on speed, deviations from typical speeds can lead to the following misclassifications:

• False negatives occur when fishing takes place at atypical speeds. A false negative means the estimation assumes that no fishing is occurring, even though fishing is actually taking place.

• False positives occur when vessels travel at speeds normally associated with fishing. A false positive means the estimation assumes that fishing is occurring, even though it is not.

The method tends to produce more false positives, but this is reduced by cross‑checking with logbook data, since fishers typically record when fishing actually occurs. As a result, unrealistic tows are removed.

Not all fishing trips have sufficient positional data for accurate estimation. Certain fisheries, such as Danish seining and gillnetting, require high data frequency, whereas bottom trawling can be estimated reliably with position intervals of up to one hour. Missing data means that the map underestimates fishing activity, particularly for some gear types. Work is ongoing to implement coverage estimates for each fishery in the mapping tool.

Limitations in methodology and data

In general, the results present a (slightly) simplified overview of fishing activity. It is important to be aware that the processing and presentation of data has its limitations. Consequently, the results on the map cannot be compared directly to the results in the tables, with the tables being more accurate.

We have adopted a method that divides the selected area into small squares. Fishing activity is then determined for each square, based on satellite recordings of the fishing vessels. Although we take the amount of fishing activity into account, all fished squares will be colored in the map, even when only a small proportion of the square is affected by fisheries.  

In addition, due to the amount of data, we have aggregated the map to a resolution with squares of approx. 300 x 500 meters, while the quantitative indicators shown in the tables are calculated for squares of approx. 60 x 100 meters. This difference in scale contributes to the higher accuracy of the numeric indicators in the tables.

The Percentage Unfished Area (PUA) in Danish fisheries may be overestimate

The table value for the Percentage Unfished Area (PUA) takes into account that a very large number of squares is only partially fished. Moreover, it is determined at a finer spatial scale than what is shown on the map. This indicator is therefore more accurate than the map.

However, it is important to be aware that the PUA estimate for periods spanning multiple years is determined as the minimum annual PUA observed. As such, the presented PUA could represent an overestimation of the actual PUA.

The Percentage Unfished Area (PUA) in international fisheries may be underestimated

Due to insufficient access to other countries' data for fishing within the mapping area, the indicators for the international fishing effort (Danish + foreign vessels) are calculated from publicly available data from the International Council for the Exploration of the Sea (ICES). This data is much less detailed on the spatial scale (rectangles of approx. 3000 x 5000 meters), which means that the indicator 'Percentage Unfished Area' (PUA) for international fisheries is significantly underestimated