Abstract and Figures

A methodology had been proposed for cross-matching visible infrared imaging radiometer suite (VIIRS) boat detections (VBD) with vessel monitoring system (VMS) tracks. The process involves predicting the probable location of VMS vessels at the time of each VIIRS data collection with an orbital model. Thirty-two months of Indonesian VMS data was segmented into fishing and transit activity types and then cross-matched with the VBD record. If a VBD record is found within 700 m and 5 s of the predicted location, it is marked as a match. The cross-matching indicates that 96% of the matches occur while the vessel is fishing. Small pelagic purse seiners account for 27% of the matches. Other gear types with high match rates include hand line tuna, squid dip net, squid jigging, and large pelagic purse seiners. Low match rates were found for gillnet, trawlers, and long line tuna. There is an indication that VMS vessels using submersible lights can be identified based on consistently low average radiances and match rates under 45%. Overall, VBD numbers exceed VMS vessel numbers in Indonesia by a nine to one ratio, indicating that VIIRS detects large numbers of fishing boats under the 30 Gross Tonnage (GT) level set for the VMS requirement. The cross-matching could be used to identify “dark” vessels that lack automatic identification system (AIS) or VMS.
Content may be subject to copyright.
remote sensing
Article
Cross-Matching VIIRS Boat Detections with Vessel
Monitoring System Tracks in Indonesia
Feng-Chi Hsu 1,2 , Christopher D. Elvidge 2,*, Kimberly Baugh 1,2 , Mikhail Zhizhin 1,2,
Tilottama Ghosh 1,2, David Kroodsma 3, Adi Susanto 4,5, Wiryawan Budy 6,
Mochammad Riyanto 6, Ridwan Nurzeha 7and Yeppi Sudarja 7
1Cooperative Institute for Research in the Environmental Sciences, University of Colorado, Boulder,
CO 80303, USA; fengchihsu@mines.edu (F.-C.H.); kbaugh@mines.edu (K.B.); mzhizhin@mines.edu (M.Z.);
tghosh@mines.edu (T.G.)
2Earth Observation Group, Payne Institute, Colorado School of Mines, 1500 Illinois St., Golden,
CO 80401, USA
3Global Fishing Watch, Washington, DC 20036, USA; david@globalfishingwatch.org
4Department of Fisheries, Faculty of Agriculture, University of Sultan Ageng Tirtayasa, Banten 42124,
Indonesia; adisusanto@untirta.ac.id
5Indonesia-Center of Excellence for Food Security, University of Sultan Ageng Tirtayasa, Jalan Raya Jakarta
Km 4, Panancangan, Cipocok Jaya, Kota Serang, Banten 42124, Indonesia
6
Department of Fisheries Resources Utilization, Faculty of Fisheries and Marine Sciences, Bogor Agricultural
University, Jl. Raya Dramaga Kampus IPB Dramaga Bogor, West Java 16680, Indonesia;
bud@psp-ipb.org (W.B.); mochammadri@apps.ipb.ac.id (M.R.)
7Kementerian Kelautan dan Perikanan, KKP Gedung Mina Bahari I Lt 5 Jl. Medan Merdeka Timur No. 16,
Jakarta 10110, Indonesia; ridwan.nurzeha@gmail.com (R.N.); abiyeppi62@gmail.com (Y.S.)
*Correspondence: celvidge@mines.edu
Received: 3 March 2019; Accepted: 19 April 2019; Published: 26 April 2019


Abstract:
A methodology had been proposed for cross-matching visible infrared imaging radiometer
suite (VIIRS) boat detections (VBD) with vessel monitoring system (VMS) tracks. The process
involves predicting the probable location of VMS vessels at the time of each VIIRS data collection
with an orbital model. Thirty-two months of Indonesian VMS data was segmented into fishing and
transit activity types and then cross-matched with the VBD record. If a VBD record is found within
700 m and 5 s of the predicted location, it is marked as a match. The cross-matching indicates that
96% of the matches occur while the vessel is fishing. Small pelagic purse seiners account for 27% of
the matches. Other gear types with high match rates include hand line tuna, squid dip net, squid
jigging, and large pelagic purse seiners. Low match rates were found for gillnet, trawlers, and long
line tuna. There is an indication that VMS vessels using submersible lights can be identified based on
consistently low average radiances and match rates under 45%. Overall, VBD numbers exceed VMS
vessel numbers in Indonesia by a nine to one ratio, indicating that VIIRS detects large numbers of
fishing boats under the 30 Gross Tonnage (GT) level set for the VMS requirement. The cross-matching
could be used to identify “dark” vessels that lack automatic identification system (AIS) or VMS.
Keywords: VIIRS; DNB; nighttime lights; VMS; IUU; boat detection; low light imaging; Indonesia
1. Introduction
The continuing decline in global fish stocks places an increasing importance on fishing vessel
surveillance to better understand individual fishing grounds and to provide a basis for enforcement of
fishing regulations and counter illegal, unreported, and unregulated (IUU) fishing activities. However,
there is no uniform and mandatory reporting on fishing boat activity or their catch.
Log books
Remote Sens. 2019,11, 995; doi:10.3390/rs11090995 www.mdpi.com/journal/remotesensing
Remote Sens. 2019,11, 995 2 of 26
recording date ranges, locations, and catch weights are required in certain jurisdictions. However,
these records are typically not submitted until after the vessel returns to its landing site and there is no
centralized system for sharing these records. Individual fishing boats can be tracked in near real time
using vessel tracks from automatic identification system (AIS) and vessel monitoring systems (VMS).
However, these systems provide an incomplete view of fishing boat activity and the data are generally
restricted in terms of access. In terms of completeness, the requirements for AIS and VMS generally
cover only the larger boats, with smaller boats able to operate “under the radar.” Indonesia has
implemented a monitoring system for fishing vessels on a voluntary basis since 2003. While formal
implementation of VMS attached on board of fishing vessel and fish carrier vessel for 30 gross tonnage
(GT) or larger has been formalized by the Ministry Regulation number 42 in 2015, which represent
approximately 10% of the domestic fishing fleet [
1
]. In the USA, VMS is required on fishing boats longer
than 19 meters and, in certain federal waters, all fishing boats are required to carry
VMS [2]
. In other
countries, such as the Philippines, there is no VMS requirement as of 2018. The International Maritime
Organization (IMO) requires that all boats 300 GT and larger carry
AIS [3]
. The vast majority of fishing
boats fall under these weight limits and are under no requirement to broadcast their location. A second
shortcoming of VMS and AIS for vessel surveillance is that operators are able to evade detection by
disabling their devices, reporting a false identification, or reporting a series of false locations. Amongst
maritime enforcement agencies, there is substantial interest in identifying “dark vessels” that lack
an accurate AIS or VMS signal. This is a tip-off of possible illegal fishing in certain fishing grounds.
The only way to identify “dark vessels” is to combine AIS and VMS with complimentary vessels
detection sources.
Certain satellite remote sensing systems can detect vessels and have potential value in fishery
surveillance. High spatial resolution optical sensors are able to detect and characterize vessels under
low cloud daytime conditions. Satellite synthetic aperture radar (SAR) systems are also able to detect
offshore vessels and have the advantage of all-weather operations [
4
,
5
]. Also, high resolution optical
imagery have been used to detect marine vessel activities [
6
9
]. The downside for these sources are
that it may take several days to access the data, there are generally fees associated with the data access
and global coverage is not currently available on a daily basis.
Data from meteorological satellite sensors have the favorable characteristics of providing global
coverage, with data open access and near real time availability. In addition, there are long term archives
for meteorological satellite data, which can be used to develop extended temporal records. The major
disadvantage of meteorological satellite data for vessel detection is the spatial resolution, which is
typically far too coarse for the detection of vessels using normal spectral bands. The exception to this
is the detection of electric lighting on boats at night.
It has been known since the 1970s that lights from fishing boats can be detected with low light
imaging data collected at night by sensors flown on weather satellites [
10
]. The low light imaging
on these sensors is designed for the detection of moonlit clouds in the visible, but also enables the
detection of electric lights present at the Earth’s surface [
11
]. One of the lighting sources detected are
fishing boats that deploy lights to attract catch. More recently, boat detection has been demonstrated
using low light imaging data collected by the NASA/NOAA visible infrared imaging radiometer suite
(VIIRS). Multiple groups have utilized VIIRS boat detection data in their research [
12
16
].
The first
VIIRS was launched in 2011 and the second in 2017 [
17
], which will help to provide more coverage per
day. In fact, the Earth Observation Group operates a near real-time VIIRS boat detection system which
produces a nightly global mapping of VIIRS boat detections (VBD), and are available for open access
download [18].
The purpose of this paper is to define a VBD cross-matching algorithm suitable for vessel track
records like AIS or VMS, and to use this to characterize the VBD product. Studies had been carried
out to match vessel track record with satellite based boat detection [
19
23
]. Most of these studies use
vessel track records to verify the vessel detection result from selected scenes of satellite imagery. To aid
satellite based vessel detection like VBD, whose records lack specific information on the boat detected,
Remote Sens. 2019,11, 995 3 of 26
track data like VMS and AIS have to be further analyzed to provide information beyond vessel name
and type. It is particularly important for fishery management to be able to answer questions like:
is a detection
an indication of fishing activity? Is it possible to identify “dark vessels” detected by
satellite but lacking track records? What percentage of fishing vessels are detected by satellite?
To investigate these questions, we conducted a cross comparison analysis of VBD and VMS
records for more than 3600 vessels spanning 32 months for Indonesia. The VMS records include
unique identifiers for individual vessels and gear type registrations. By cross matching the VMS tracks
with VBD it is possible to calculate the match rates and average VIIRS radiance for individual vessels.
These results are aggregated for all the vessels having a specific gear type registration to answer the
question on the types of fishing vessels detected by VIIRS.
In addition to cross matching, we also analyze VMS tracks for the status of the vessel at the time
of recording. The VMS tracks are classified into four activity types by their location, speed and heading
changes: landing, transit, stationary, and maneuvering. Such efforts help to answer the question on
whether a VBD can be interpreted as “fishing.” Separate match rate summaries are calculated for
fishing and transit activity types.
Clearly, any new data source on vessel detection needs to be thoroughly examined and compared
to other available data sources. Otherwise the users will not know how to interpret the data or develop
standard operating procedures for their use. Our intention is to provide results on VBD in reference to
a widely recognized vessel track data source.
2. Methodology
2.1. Data Collection
2.1.1. VMS
Under a cooperative research agreement with the Global Fishing Watch (GFW), we obtained
32 months of VMS data collected by the Indonesian Ministry of Marine Affairs and Fisheries (MMAF
a.k.a. KKP). This consists of records for Indonesian fishing vessels larger than 30 GT from January
2014 to August 2016. The VMS records comprise 15 gear types with more than 3600 distinct vessels.
The typical gap between vessel track records is about an hour. Here the term “track record” refers
to the location report of the vessel at the time of reporting. The summary of gear types included
in the database is shown in Table 1. The records include transmitter number, timestamp in UTC,
latitude/longitude, registered gear type, vessel size and other useful information. There are nearly
29 million VMS records, with roughly 20 million track-hours. For easier reference to the gear types,
abbreviations are devised as shown in Table 1and used throughout this work.
Table 1. Summary of gear types in the vessel monitoring system (VMS) database.
Gear Type Abbr. # Vessels Average GRT Recorded Time
Bahasa English (Std. dev.) (Hours) 1
Pukat cincin pelagis kecil Small pelagic purse seiner PCK 1085 99.9(50.0) 3,357,444
Bouke ami Stick-held squid dipnet BA 502 71.2 (28.2) 1,441,394
Rawai tuna Longline tuna RT 428 113.5(84.7) 1,806,814
Pukat cincin pelagis besar Large pelagic purse seine PCOB 417 129.9(55.9) 1,936,373
degan satu kapal with one ship
Pengangkut Carrier 3P 386 300.8(539.0) 1,825,319
Pukat ikan Trawler PI 205 210.9(128.3) 6777
Pancing cumi Squid jigging PC 195 107.8(41.3) 1,377,622
Jaring insang oseanik Oceanic gill net JIO 172 142.6(106.1) 742,990
Jaring liong bun Shark gillnet JLB 139 48.7(13.5) 82,034
Rawai dasar Basic longline RD 132 77.9(59.4) 945,630
Hand line tuna 2Hand line tuna HLT 69 98.46(43.8) 88,705
Huhate Pole and line H 67 63.9(18.8) 205,372
Pukat udang Shrimp trawl PUD 42 152.8(34.1) 0
Remote Sens. 2019,11, 995 4 of 26
Table 1. Cont.
Gear Type Abbr. # Vessels Average GRT Recorded Time
Bahasa English (Std. dev.) (Hours) 1
Pukat cincin grup pelagis Small pelagic purse PCGK 12 83.8(18.9) 0
kecil seine group
Pancing ulur Hand line PUR 4 107.5(34.1) 0
1
Record time: hours are calculated with record interval less than 2 h, excluding erroneous records.
2
Recorded
in English in track database.
3
Carriers are vessels which collect catches from fishing vessels.
Gear types
with more than 1 million hours recorded in the database are in boldface.
Gear types Bubu (Fish trap) and
Pukat cincin grup pelagis besar (Large pelagic purse seine group) were found in the vessel list, but do not
have any record in the VMS track database.
2.1.2. VBD
The VBD data used in this study were downloaded from the EOG website [
18
] for the
corresponding time range, i.e., from January 2014 to August 2016. The VBD records were filtered
to a rectangular area covering the Indonesian Fishery Management Zones (Wilayah Pengelolaan
Perikanan, WPP) as shown in Figure 1. The VBD data used in this study were produced by the V23
algorithm. VBD records were also imported to the same database as the VMS records for further
processing. The VBD data were provided in the form of CSV files, with data from columns listed in
Table 2used in the study.
Figure 1.
Bathymetry map of Indonesian waters with Indonesia Wilayah Pengelolaan Perikanan
(WPP). WPP regions courtesy to Ministry of Marine Affairs and Fisheries (KKP) [24].
VBD from 2014–2016 being a dataset derived from observations of a single polar orbit satellite
with nominal swath of 3000 km, can cover the globe daily, with minimal coverage of 1 at the equator,
and increased coverage closer to both poles. Indonesia usually gets 1 coverage, and 2 coverages in the
orbit overlap area. The band used for detection in VBD is VIIRS day night band (DNB), with a nominal
pixel footprint of 742 m by 742 m. VBD also adopts data from VIIRS Nightfire which was also a product
of EOG [25], to prevent gas flares of oil platforms from being recognized as lit vessels [16].
Remote Sens. 2019,11, 995 5 of 26
Table 2.
Attributes provided by visible infrared imaging radiometer suite boat detections (VBD) used
in this study.
Name Explanation Unit
ID_Key Unique key for each VBD record Unitless
Lat_DNB Latitude of the center of DNB pixel Degrees
Lon_DNB Longitude of the center of DNB pixel Degrees
Date_Mscan Timestamp in UTC at the middle of the scan line Unitless
Rad_DNB Radiance of the DNB pixel nW/sr/cm2
QF_Detect Quality flag (type) of the VBD detection Unitless
Note: See VBD web site [18] for details on header definitions.
2.2. Vessel Location Prediction
To match VMS to VBD, the vessel location at the time of satellite overpass has to be known. This is
achieved by interpolating the vessel location between neighboring records at the time of anticipated
satellite overpass time. The location interpolation is shown as
Xt=X1+X2X1
T2T1
×(TtT1)(1)
Yt=Y1+Y2Y1
T2T1
×(TtT1), (2)
where
Xt
and
Yt
are coordinates of the vessel at the satellite overpass time
Tt
.
X1
and
Y1
are vessel
coordinates in VMS records immediately prior to
Tt
. Likewise,
X2
and
Y2
are vessel coordinates in
VMS records immediately after
Tt
. This can also be re-written to do extrapolation of vessel location at
the beginning and ending of vessel tracks. Such process are called vessel location prediction in this
study. The overall process is shown in Figure 2.
Before performing vessel location prediction, the VMS track record needed to be cleaned to
remove records with erroneous timestamps or coordinates. Records that indicate moving speed larger
than 30 knots were also neglected.
VMS track series were processed by UTC date. The record that was nearest to local midnight
was selected, whose coordinate and timestamp were used as seed to begin the search for the nearest
satellite overpass time, here referred as the predicted time.
A program, VIIRS overpass predictor, was developed to find the closest satellite overpass time for
a given location and time. The orbital location of satellite at any given time can be propagated with
SGP4 model [
26
] and proper two line element (TLE) [
27
]. TLE provides all the necessary parameters
describing the satellite orbit at a specific time. TLE is retrieved from CelesTrak which usually being
updated daily [28].
Since TLE was frequently updated, the program finds the TLE closest to the seeding time, and
it propagates satellite location with 10 min interval until the seeding location passed the scanning
plane and is visible, i.e., the shortest distance from target to satellite path is within the scanning swath.
Then it reverses the direction of propagation with 1/10 of the previous time step until the seeding
location passed the scanning plane again. The time step is then reversed with 1/10 of its current setting.
The process is repeated until the change of angle between the satellite to the visible observer and the
scanning plan is less than 0.0001 degrees for the current time step. The time step is then reset and the
next loop is initiated with seed time advanced by 60 min from the current predict time to search for the
next possible overpass. When the given coordinate is within the orbital overlap zones, it is possible to
find multiple predict overpass times. The process is continued until the seed time is 24 h advanced
from the first seed time. By verifying the predicted time with VBD records, the precision of the VIIRS
overpass predictor is determined to be ±2 s.
Remote Sens. 2019,11, 995 6 of 26
Figure 2. Flow chart of vessel location prediction at satellite overpass.
The predicted VIIRS overpass time is then used to interpolate the vessel location between
immediate neighboring VMS track before and after it. If the predicted VIIRS overpass time is beyond
the ending or starting record of the VMS track series, extrapolation is adopted. The process is skipped
if the predicted VIIRS overpass time is 2 h away from any nearest VMS track records. The calculated
record with predicted VIIRS overpass time and interpolated vessel location is referred as predicted
record. Such records are inserted into the VMS database and treated as part of the VMS track series.
2.3. VMS Record Status Classification
VMS tracks with predicted records inserted are then classified by their type of activity. In this study
we tried to separate the type of activity into maneuvering (M), transit (T), stationary (S), and landing (L).
Briefly speaking, M records were those with lower speed and larger heading change, while T records
were those with higher speed and smaller heading change. S and L records were those not moving
for a period of time, with L records were found close to recognized landing sites. We specifically call
vessels in M and S as being fishing (F). There are a number of different methods developed to perform
status classification of track records [
29
]. Here the method proposed by de Souza et al. [
30
] for long
line vessels using an R package called adehabitatLT [31] was chosen.
Status classification as shown in Figure 3was applied on one vessel within a calendar year at
a time. If the number of track records within a calendar year was less than 50, the track will not
be subjected to classification. The adehabitatLT package requires the track to be evenly separated
temporally. The VMS records were not strictly 1 h apart, hence it had to be normalized to a regular
hourly interval before classification.
The coordinates were reprojected to corresponding universal transverse mercator (UTM) zones in
accordance to the average latitude and longitude of the selected annual track to have the calculation
carried out in meters instead of degrees. Although vessels can cross multiple UTM zones, the errors
caused were assumed negligible in the scope of this study. This was because fishing vessels usually
take regular routes and are unlikely to cross more than two UTM zones.
Remote Sens. 2019,11, 995 7 of 26
The process of applying adehabitatLT can be simplified in two stages. First, the annual track was
cut into segments. The maximum likelihood method gives the most likely number of segments in the
track, which describes how many times a vessel changed its pattern of maneuvering in the given track.
Then for each segment, we found the most likely model from the predefined velocity model for each
segment [31].
The velocity model in this study was defined as a series of Gaussian distribution with mean = 2
(km/h)
×
n
where
n
=
1
to
10, and standard deviation as 2.5 (km/h). For our purpose, segments with
velocity model lower than or equal to model 5, i.e., velocity normal distribution of mean velocity of
10 km/h
with standard deviation of 2.5 km/h, are determined to be maneuvering (M). Fast moving
segments were deemed in transit (T). If slow moving segments have an average cosine of turning
angle between records larger than 0.8 or smaller than
0.8, i.e., angle smaller than approximately
36.8 degrees, then the segment was also considered as in transit.
Besides checking by segment, the track was also checked record-wise for stationary records. First,
the track was scanned for stationary records. If any of the four consecutive records moved less than
100 m, their status was set to stationary (S). If the whole track was stationary, no further process
was applied, and the whole track was marked as stationary. The distance to the nearest coast of
these stationary records are then determined by querying the NASA coastal distance database [
32
].
The status of records which are within 2 km from the nearest coast were overridden to landing (L),
and a possible landing site was written the the record by referencing the list of known landing sites,
which is extracted from the MMAF website [33].
Segment-wise and record-wise results were then consolidated and written into normalized track
record, which was then de-normalized back to its original time stamp and written into the database.
The same algorithm was applied to all gear types, as we found the output to be satisfactory within the
scope of this study.
Figure 3. Flow chart of VMS track status classification.
2.4. VBD Cross Matching
The threshold of matching predicted VMS record to VBD was set to be within 700 m and 5 s.
The reason to set the distance threshold at 700 m was in consideration of the VIIRS-DNB pixel footprint
size, while a temporal threshold of 5 s was because the precision of overpass prediction was
±
2
s. Such a threshold was considered to provide enough tolerance. If a VBD match was found for
Remote Sens. 2019,11, 995 8 of 26
any predicted record, the matched VBD record ID along with the distance and temporal difference
was recorded.
There were cases when a VBD record was found with similar coordinate in neighboring orbits.
Being able to add temporal discrimination in matching VMS and VBD gave us the ability to match the
exact VBD record within orbit overlap areas.
3. Results
3.1. Status Classification Result
Examples of classification results are shown in Figure 4. Figure 4a shows the track of a small
pelagic purse seiner vessel (PCK) from January to August 2016. It is clear that the vessel has a rather
simple pattern with single landing site of Pekalongan and focused fishing ground in Makassar Strait.
The red dots indicate VMS records that are recognized as in transit (T). Blue dots are those in fishing
(F), i.e., maneuvering (M) or stationary (S). Green dots represents when the vessel was near the coast,
and supposed to be landing at Pekalongan in this case. The vessel also appears to be stationary near
Saburu Island. Figure 4b shows the track of a squid jigging vessel (PC) also from January to August
2016. The vessel travelled for weeks from Jakarta to Arafura Sea for fishing in the two most popular
squid jigging sites. It appears that the vessel got its supply from Merauke while operating continuously
in the Arafura Sea for months. It also indicates that the vessel had stopped at Jampea Island. It would
be reasonable to assume that it also landed its catch in Merauke as well. The vessel returned to Jakarta
after its long voyage.
Figure 4.
(
a
) Pelagic purse seiner vessel (PCK) track classification result of VMS track in 2016 from
January to August. The vessel had been fishing in Makassar Strait, and landed in Pekalongan and
Saburu Island. (
b
) Squid jigging vessel (PC) track classification result of VMS track in 2016 from January
to August. The vessel had been fishing in the Arafuru Sea and the Banda Sea, and landing in Merauke,
Jampea Island, and Jakarta. Base map provided by Open Street Map [34].
Remote Sens. 2019,11, 995 9 of 26
Figure 5visualizes the distribution of velocity and heading changes for moving VMS records,
i.e., transit and maneuvering. It is clear that vessels in transit travelled straight for most of the time,
with velocity mostly found at 10 km/h. For those in maneuvering, they often changed heading with
speed around 1 to 2 km/h and wider spread of heading changes.
Figure 5.
Velocity and heading change for all vessels in the VMS database. (
a
) For records in transit
and (b) for records in maneuvering. Markers are colored by the number of VMS records.
3.2. Cross-Matching Result
Here the result of cross-matching is displayed. Figure 6shows the density map in 0.1 degree cells,
with VMS records marked as fishing (F) in pink and red, and cells with matched VBD record as green,
for the top 6 gear types whose recorded time is larger than 1 million hours as shown in Table 1. Carrier
(P) is excluded despite having more than 1 million hours of recorded time for it is not a fishing vessel.
The gear type with most recorded hours next to P, Basic longline (RD), is included instead.
It is intriguing to see how each gear type prevails distinctive water. While stick-held squid dipnet
(BA)/squid jigging (PC)/small pelagic purse seiner (PCK)/RD vessels focus mainly in the Java sea
and the Arafura sea, large pelagic purse seine with one ship (PCOB) and longline tuna (RT) spreads
widely into the Indian Ocean. The later two gear types were larger in gross tonnage as shown in
Table 1, which implies they were better adapted to rough waters and can stay in the oceans longer
without resupply.
Figure 6shows the density map also in 0.1 degree cells of VMS records with matched VBD
record found, with Figure 6f showing the VBD records with detection quality flag (QF_Detect) of 1,
i.e., high quality
boat detection, from 2014 to 2016 for reference. For those gear types which show
small area as green indicates having fair chance of successful cross-match, i.e., finding VBD record
match for the predicted VMS record. BA, PC, and PCOB all have similar distributions for VMS fishing
records and their VBD matches. PCK only shows similar patter in WPP (Fishery Management Zone)
711, 712 and 713. RT although shows wide spread of VMS fishing records in south of Java, west of
Sumatra, and the Arafura Sea as shown in Figure 6e, it is only occasionally matched, with most of
them happening in the Arafura sea. RD operation was heavily found in the Arafura sea, and was
rarely being matched by VBD records.
Figure 7shows the gear type composition of VMS record with VBD found for each WPP region.
PCK is dominant in WPP 571, 572, 711, 713, 715, 716, and seen in WPP 573, 712, 714, 717. PCK is seen
in all 11 WPP regions with 10 having substantial to dominant proportions. PC has absolute weight
in WPP 718, while WPP 718 also has the largest amount of VMS match records. BA matches mostly
exist in WPP 711 and 712, and some in WPP 718. PCOB is most visible in WPP 573, while RT show
substantial proportion in WPP 717 and 714. However, most vessels were detected by VBD in WPP 572,
711, 712, 713, 573, and 718. The rest of the WPPs barely have any matches.
Remote Sens. 2019,11, 995 10 of 26
Figure 6.
Spatial distribution of VMS fishing records (pink and red) with matched VBD detection
(green) for major gear types in 0.1 degree cells.
Figure 7.
Composition of major gear types for matched VMS records in each WPP region. The size
of the pie chart indicates the total number of matched VMS records for major gear types included,
note the size is not strictly proportional to the actual number. WPP regions courtesy to KKP [24].
Since polar orbiting satellites like Suomi National Polar-orbiting Partnership (Suomi NPP) which
carries VIIRS only observes a given location in limited time range, it is necessary to evaluate how
much fishing activities are accounted for in VBD. As shown in Figure 8, VIIRS overpass window only
covers 22:00 to 4:00 in local time for Indonesia, and is most frequent at 1:00.
Here we overlay the temporal distribution of records in fishing status for BA, PC and PCK,
which were the most dominant gear types as shown in Figure 7, to show their diurnal patterns of
fishing and stationary in Figure 8. It is clear that BA, PC and PCK tended to be in fishing status at
24:00 to 3:00 local time, with peak at 1:00. The fact that the time range when vessels of these gear types
Remote Sens. 2019,11, 995 11 of 26
were at fishing status coincides with the VIIRS overpass window suggests that VIIRS observation can
account for most fishing activities at night.
Figure 8.
Distribution of normalized occurrence for visible infrared imaging radiometer suite (VIIRS)
overpass over Indonesia, and PC and PCK in fishing status.
3.3. Match Rate
Once it was possible to match predicted VMS records, i.e., vessel location at the time of predicted
VIIRS overpass, with VBD, it was natural that we would like to quantify the performance of
cross-matching for each gear type. Table 3displays details of match rate calculations for each gear type
available in the VMS database. Small pelagic purse seiners (PCK) accounted for 27% of the matches.
The raw match rate for certain gear type (
Rraw,i
) was the simplest way to evaluate the cross match
performance for gear type i. It was calculated as
Rraw,i=Cm,i
Cp,i
, (3)
where
Cm,i
is the match count, and
Cp,i
is the prediction count, for gear type
i
. Below we drop the
i
sign to simplify the formulas. The prediction count is the sum of interpolated vessel location records
within the time period covered by the dataset.
However, the number of predicted vessel locations should be adjusted for the scenario when
heavy cloud cover blocks vessel detection. Rather than screen out all pixels having cloud using the
VIIRS on/off cloud mask, we wanted to include a broader range of cloud conditions, as we know from
visual inspection that VBD produces detections even in the presence of some cloud cover. To address
this issue, we developed a cloud transparency index based on the VIIRS imaging band five. I5 observes
longwave infrared radiation at 10.5 to 12.4
µ
m [
35
]. Over the warm waters surrounding Indonesia,
clouds are colder than the underlying ocean surface, and will have lower radiance values in the I5
imagery. To derive the cloud transparency index, we generated a histogram tallying the number
of VBD detections for 600 narrow I5 radiance bins. Then, as the reference, we made a companion
histogram tallying the number of ocean background I5 pixels for the same radiance bins. By dividing
these two histograms, a clear linear relationship emerged, depicting cloud transparency as a function of
I5 radiance. As the I5 radiance increased, the number of VBD detections increased, indicating increased
cloud transparency. At an I5 radiance of 1.5 or less, the cloud was effectively opaque, blocking all vessel
detections. I5 radiances in the range of 7 to 8 corresponded to the highest level of transparency, largely
cloud-free conditions. We then converted the VBD/background ratio values to a cloud transparency
index by rescaling from 0 to 1 (Figure 9). Thus, we were able to take the I5 radiance and calculate
Remote Sens. 2019,11, 995 12 of 26
cloud transparency. We used this capability to adjust the number of predicted VMS locations prior to
calculating match rates.
Figure 9.
VIIRS I5 radiance and cloud transparency index. I5 values between 1.5 and 8 were included
in the regression. I5 radiances higher than 8 will have a transparency index of 1 (clear sky), and lower
than 1.5 will have a transparency index close to 0 (opaque sky).
We corrected for the effect of clouds on VBD using this atmospheric transparency as correction
factor as
Cp_adj =
r
Cp(r)
O(r), (4)
where
r
is the I5 radiance of each bin,
O(r)
is the cloud transparency index in regards to I5 radiance,
and
Cp_adj
is the adjusted prediction count. With the adjusted prediction count (
Cp_adj
), the adjusted
match rate can be calculated as
Radj =Cm
Cp_adj
. (5)
We were particularly interested to know the chance of a vessel to be detected by VBD in particular
operation status. Such specialized match rate is defined as
Radj (F) = Cm(F)
Cp_adj (F), (6)
Radj (T) = Cm(T)
Cp_adj (T), (7)
where Fand Tdenotes the status: “fishing” or “transit”.
Given all the information, we can also calculate the probability of a VBD detection being in fishing
status as
P(F) =
i
Cm,i(F)
i
Cm,i(F) +
i
Cm,i(T), (8)
where
i
denotes each gear type. Note that we only account for matches in the status of fishing and
transit in the denominator. That is because vessels in landing status were not considered to be in
operation, and VBD could be influenced by city lights which causes false detection. In this study,
P(F)
is determined to be 96.42%.
Remote Sens. 2019,11, 995 13 of 26
Table 3. Match rate for each gear type.
Gear Type Matches 1Predictions 2Rater aw 3Rateadj (F)4Rat ead j(T)5
Hand line (PUR) 108 259 41.7% 82.8% 5.8%
Stick-held dipnet (BA) 40,999 101,111 40.5% 80.8% 10.4%
Squid jiggiing (PC) 35,001 77,431 45.2% 78.3% 9.1%
Hand Line Tuna (HLT) 3691 10,961 33.7% 72.7% 5.8%
Small pelagic purse seiner (PCK) 59,638 242,727 24.6% 58.9% 7.7%
Large pelagic purse seine with one ship (PCOB) 18,225 121,535 15.0% 32.9% 8.3%
Pole and line (H) 114 12,998 0.9% 10.9% 0.5%
Longline tuna (RT) 3159 72,817 4.3% 10.2% 4.3%
Trawler (PI) 1069 28,203 3.8% 8.3% 5.6%
Oceanic gill net (JIO) 1259 49,073 2.6% 6.3% 2.5%
Shrimp trawl (PUD) 247 11,643 2.1% 6.6% 2.5%
Carrier (P) 293 63,541 0.5% 6.2% 0.9%
Shark gillnet (JLB) 82 12,630 0.7% 1.2% 0.3%
Small pelagic purse seine group (PCGK) 4 1047 0.4% 0.4% 2.2%
Basic longline (RD) 104 50,447 0.2% 0.3% 0.2%
1Cm
: Count of matched records.
2Cp
: Count of predicted records.
3Rraw
: Raw matching rate.
4Radj (F)
: Adjusted matching rate for fishing status.
5Radj (T)
: Adjusted matching rate
for transit status.
Remote Sens. 2019,11, 995 14 of 26
3.4. Average Radiance and Match Rate
Through VMS/VBD cross-matching it was possible to identify the gear types that are widely
using lights to attract catch and those that are not. It is possible that vessels of the same gear type may
differ in both wattage and the manner in which lights are deployed. There is evidence for regional
differences in lighting for small pelagic purse seiner (PCK). As shown in Tables 4and 5, PCK in
different regions exhibit different average radiances and hence may have different raw match rate.
The results in Table 4shows that PCK in the Natuna Sea and the Makassar Strait has near 45% chance
of being detected by VBD, while those in Malaku Sea have less than 1%. Table 5shows that PCK used
brighter lights and were more detectable in WPP 711 near the Natuna Sea as well as WPP 712 and 713
near the Makassar Strait. PCK operating in the Maluku Sea in WPP 715, 716 and 717 did not seem to
use light at all. We suspect such difference was caused by the extensive use of fish aggregating devices
(FADs) in WPP 715, 716, and 717. By attracting catch with FADs, fishing vessels only need enough
lighting for operating, eliminating the use of excessive lighting.
FADs are floating objects with extensions hanging in the water that collect sea-life, such as that
attract pelagic species including tuna, mackerel, scads, sardines, etc. There are two common types of
FADs. Pontoon or box type is made of steel, and raft type made of bamboo. Attractors hanging below
the surface of water can be made of coconut leaves, nipah leaves, and pinag leaves [
36
]. These devices
are anchored to prevent from drifting. Fishes will tend to be attracted by the hanging attractors or
smaller species aggregated around the device, therefore fishermen do not require strong lights to
attract catches or baits. FADs are known to be a old and common practice in Indonesia, enabling
fishermen to use minimal effort for maximal catch [36].
Table 4. Match rate for pelagic purse seiner vessel (PCK) in different regions.
Region WPP # VMS Counts Raw Match Rate
Match Predicted
Natuna Sea WPP 711 19,982 45,341 44.10%
Makassar Strait WPP 712
WPP 713 31,827 71,456 44.50%
ine Maluku Sea
WPP 715
WPP 716
WPP 717
100 28,825 0.35%
Note: data from January 2014 to August 2016.
Figure 10 shows the relationship between average DNB radiance and match rate for each vessel
and gear type. Stick-held dipnet (BA) clearly shows a fairly stable average radiance and match rate
among all vessels. Squid jigging (PC), however, has a wider spread on average radiance, but the match
rate is more stable in comparison.
As shown in Figure 10c, PCK clearly shows two distinct populations with different use of lighting.
While one group has a comparable average radiance and match rate with BA, the other group uses
almost no lighting and resulted in lower match rate. This result corresponds to
Tables 4and 5
that
the group of points with lower radiance and match rate is mostly comprised by PCK operating in
the Maluku Sea. The reason that purse seiners operating in Maluku Sea fishing ground are seldom
detected by VBD can be attributed to the use of FADs instead of on-board lighting to attract their catch,
as reported by Natsir and Atmaja [
37
]. Another explanation observed in the field survey conducted by
the second author shows that almost all PCK vessels located in the Java Island carried strings of metal
halide bulbs above deck as shown in Figure 11, while those in the Maluku Sea region utilizes small
number of submersible lights to attract their catch.
For large pelagic purse seiner with one ship (PCOB) vessels shown in Figure 10d, although some
vessels had match rate higher than 0.4, most vessels were seldom detected by VBD with overall very
Remote Sens. 2019,11, 995 15 of 26
low light usage as suggested in Table 5and Figure 12. Most PCOB were found operating in the Indian
Ocean near West Sumatra and South Java, which are WPP 572 and 573 as shown in Figure 6. There
is high possibility that PCOBs also frequently utilize FADs to attract their main catch, which is tuna,
hence result in lower light usage. Most longline tuna (RT) vessels use very little lighting, and are
seldom detected by VBD as shown in Figure 10e. Here we include hand line tuna (HLT) vessels which
is one of the gear types with highest match rate instead of carrier (P) which has raw match rate less
than 1% (see Table 3) in Figure 10f. Although the VMS database only includes a small number of HLT
vessels, most were frequently detected by VBD and utilized a decent amount of lighting.
Figure 10.
Scatter plot of average radiance and match rate by vessel by gear types. Carrier (P) is
excluded due to too low match rate. Hand line tuna (HLT) is added instead for it having substantially
high match rate.
Remote Sens. 2019,11, 995 16 of 26
Figure 11.
Photos taken during field survey in Indonesia. (
a
) Purse seiner at Muara Angke Port,
Jakarta. Equipped with 60 bare 1500 watt metal halide bulbs, and 24 shielded bulbs pointing into the
water. (b) Closeup look of a 1500 watt metal halide bulb.
As displayed in Table 5, the average radiance for all gear types are listed, with PCK further
split into three regions. By viewing the overall average radiance, handline (PUR) vessels use the
brightest light albeit comprised by only small number of vessels. Most gear types were relatively
stable throughout the years, except RT had grown noticeable brighter. This could indicate that RT was
adopting fishing with lights beginning at 2015.
Table 5. Average radiance of gear types.
Gear Type Region Average Radiance 1
2014 2015 2016 2All
Handline (PUR) 126.50 126.50
Handline tuna (HLT) 109.63 127.24 95.22 104.61
Squid jigging (PC) 78.36 99.37 89.73 89.52
Stick-held squid dipnet (BA) 68.87 70.58 72.13 70.78
Small pelagic purse seiner (PCK) All 39.47 40.48 45.45 42.23
Natuna Sea 48.46 46.16 52.85 48.87
Makassar Strait 43.63 46.64 40.61 46.50
Maluku Sea 4.48 3.09 3.48 4.24
Shark gillnet (JLB) 41.55 19.93 25.60 26.44
Longline tuna (RT) 1.74 26.08 35.61 24.01
Carrier (P) 30.17 5.42 14.19 23.37
Large pelagic purse seine with one ship (PCOB) 4.87 4.19 4.76 4.58
Basic longline (RD) 3.24 3.30 4.48 3.43
Oceanic gillnet (JIO) 4.51 2.61 1.34 3.27
Pole and line (H) 3.00 2.97 3.08 2.99
Trawler (PI) 2.67 4.17 – 2.72
Small pelagic purse seine group (PCGK) 2.48 1.87 2.32
Shrimp trawl (PU) 1.33 1.34 1.33
1
Unit: nW/sr/cm
2
.
2
Till 2016 August. Note: regional names and numbers are in italic.
Remote Sens. 2019,11, 995 17 of 26
Figure 12. Average radiance of gear types.
The increasing average radiance level has a positive correlation to the match rate as shown in
Figure 13. PC and BA were among those gear types with best match rate and highest radiance, due to
the application of high wattage lamps to attract squid during operation. Moreover, the fishing vessel
(PC and BA) only had short movement (500–700 m) within the fishing ground, making the matched
VBD record substantially representative for the general location of the vessel during the day.
The handline (PUR) and handline tuna (HLT) did not commonly use lights to attract their catch.
The fisherman of HLT and PUR use lights to catch squid and other fish as the main natural bait for
tuna fishing. The number and wattage of lamp types on HLT and PUR fishing vessel varies. The bare
metal halide bulbs of 1000 and 1500 watts were commonly used as attractant lamp. Boats were usually
equipped with 10 to 20 lamps, evenly installed on both sides of the vessel [
38
,
39
]. Operations of PUR
and HLT to catch bait are similar to PC with limited movement in the same fishing ground.
Figure 13. Relationship of average radiance and raw match rate by gear type
Figure 14 shows the map of average DNB radiance for VMS records with matched VBD record.
By only taking records with best quality (VBD quality flag = 1), we are confident in accounting
observations without additional uncertainties. Furthermore, the map only shows cells with more than
five records to ensure the map represents repetitive behaviors. It is clear that vessels within the Java
Sea and the Arafura Sea were much brighter than those operating in the Indian Ocean. It is suspected
the radiance difference was related to FADs usage in deeper waters in the Indian Ocean, and the
differences in light requirement for PCOB operations which is more dominant in the Indian Ocean.
However, the exact reason remains to be investigated.
Remote Sens. 2019,11, 995 18 of 26
Figure 14.
Map of average day night band (DNB) radiance for matched VMS records with VBD quality
flag (QFQF_Detect) = 1 in 0.1 degree cell (only showing cells with more than five records)
3.5. VMS and VBD Record Uniqueness
Table 6shows the summary of cross-matching by comparing the matched and missed records for
both VBD and VMS. The records were limited within WPP zones to ensure fair comparison. For VMS,
20% of the predicted records were matched by VBD. The missed predicted VMS records can be due to
non-fishing activities, or fishing without heavy lights.
For VBD, only 6.72% of the records were matched by predicted VMS records. This implies that
a huge portion of lit marine activities took place at night were not accounted by the VMS. The majority
of these VBD records can be contributed by smaller vessels which are not required to carry VMS. These
unique VBD records can also be an indication of “dark vessels”. It is clear that the matched record
number for VMS was larger than VBD. Such was caused by multiple VMS being matched to the same
VBD record. This is elaborated in Section 3.6.
Table 6. Match summary for records within Indonesian WPP.
VBD 1VMS 2
Count Percent Count Percent
Match 143,917 6.72% 158,119 19.99%
Miss 1,996,168 93.28% 632,988 80.01%
Sum 2,140,086 100% 791,107 100%
1
Accounting for data from January 2014 to August 2016 and all QF_Detect flags.
2
Only count
predicted records.
3.6. Multiple Matches
It is possible for one VBD pixel to match to more than one VMS vessel. However, this was
infrequent, with 96% of the matches formed with a single VBD and single VMS boat. For the five major
gear types (not including carrier (P)), only 5800 involved multiple VMS vessels out of 152,913 matches
within Indonesia WPPs. In some cases the multiple matches can be due to one boat using lights
to attract catch and the other VMS vessel in transit. Another possible scenario could be that there
were multiple VMS vessels in close proximity to exploit a particularly rich fishing ground. Figure 15
shows the histogram of the count of VBD records ranking by the numbers of VMS matches associated
for major gear types, with the highest one having 54 VMS matches. It is clear that VMS records of
stick-held squid dipnet (BA) and squid jigging (PC) vessels have the highest chance of being matched
to the same VBD records.
Figure 16 shows the extreme example of 54 VMS records matching to a single VBD record.
The scene was found on 22 March 2013 in the Arafura Sea about 125 km south–west from Merauke.
Remote Sens. 2019,11, 995 19 of 26
It is a known popular squid jigging ground. Within all the matches, 40 are PC, while eight were hand
line tuna (HLT) vessels, two BA, and one hand line (PUR). It is a potential example of the existence
of fishing vessel sharing a smaller group or even one light vessel. This was possibly due to their
tendency to fish in smaller fishing grounds with a large number of vessels, with shorter movements
during fishing.
Figure 15.
VBD multi-match histogram accounting only matches with fishing VMS records within
Indonesia WPPs for major gear types.
Figure 16.
Map of fifty-four vessels matched to single VBD record found in the Arafura Sea,
about 125 KM
south-west from Merauke. Forty of them were squid jigging (PC) vessels with the
rest composed of stick-help squid dipnet (BA) and handline tuna (HLT). Track of the same UTC day
was plotted with different color for each vessel. The VBD record is marked with a star. The right half of
the map shows a magnified view of the tangled vessel tracks. With the DNB image as the base layer, it
can be clearly seen how bright the cluster of vessels are. Basemap provided by OpenStreetMap [34]
3.7. Matched QF_Detect
VBD provides QF_Detect to further discriminate detection based on their quality, here in Table 7
shows a break down of QF_Detect of found VBD match over the three annual periods. In practice, VBD
records with QF_Detect of 1, 2, 3, 8, and 10 were regarded as possible fishing activities. The distribution
of QF_Detect for matched VBD records followed that of all VBD population as shown by quoted
Remote Sens. 2019,11, 995 20 of 26
percentages in Table 7, indicates the chance of VBD to be matched by a VMS record is comparable
regardless of its QF_Detect state.
Table 7. Number of matched VBD records break down by quality flag (QF_Detect).
QF_Detect Explanation Matched VBD counts All VBD Counts 1,2
2014 2015 2016 1Total
1 Boat 35,218 37,141 46,200 118,559 (74.9%) 1,420,332 (71.2%)
2 Weak 4098 5102 6069 15,269 (9.6%) 220,523 (11.1%)
3 Blurry 2827 2732 2497 5556 (3.5%) 59,491 (3.0)
8 Recurring light 3669 8214 4615 16,498 (10.4%) 218,707 (11.0%)
10 Weak and blurry 626 916 845 2387 (1.5%) 27,354 (1.4%)
Sum 46,435 54,105 60,206 158,269 (100.0%) 1,994,772 (100.0%)
1
Till 2016 August.
2
Within Indonesia WPP. Note: see VBD web site [
18
] for details on quality flag
(QF) definitions.
3.8. Matched VMS Status
Here we break down the percentage of status for predicted vessel locations having VBD matches.
As shown in Figure 17, it is clear that most matches were found when the vessel is in the status of
fishing or stationary. 67% of longline tuna (RT) matches were found in fishing status, while other major
gear types all have more than 90%. RT with substantial matches found in transit status indicate that RT
vessels used a significant amount of light while being recognized as in transit. This can be due to lights
being used when longliners deploy snoods with baited hooks while moving in relatively
higher speed.
It is also worth noticing that small pelagic purse seiner (PCK) and stick-held squid dipnet (BA)
were often bright when in the status of stationary than maneuvering, unlike other major gear type
spending more lit time in maneuvering. This could also be attributed to how and what vessels of those
gear types are fishing.
Figure 17. Proportion of status for matched VMS records for major gear types.
Remote Sens. 2019,11, 995 21 of 26
4. Discussion
Our study has four objectives: (1) to identify the gear types that are commonly detected by VIIRS
in Indonesia, (2) to rate the probability that a boat is fishing if it is detected by VIIRS, (3) to determine
the probable gear type for a VBD in Indonesia fishery management zones (WPP), and (4) to determine
if different styles of lighting can be discerned. To address these objectives we combined two types of
data. The reference data are 32 months of VMS data supplied by the Indonesia Ministry of Marine
Affairs and Fisheries (MMAF). The subject data are VIIRS boat detections from the same 32 months.
Out of 3683 VMS equipped vessels, 2632 had at least one VBD match.
The methodology can be divided into two parts. The first is an algorithm for cross-matching
VIIRS boat detections with VMS tracks. The process involves predicting the probable location of
a VMS equipped vessel at the time of each VIIRS data collection using an orbital model. The probable
vessel location is interpolated between the two immediate neighboring VMS records found before and
after the predicted VIIRS overpass time. Matches are confirmed if the VIIRS has a vessel detection
within 700 m and 5 s of the predicted location. The second part involves algorithm to segments and
classifies VMS records into landing, transit and fishing activity types based on their location, velocity
and change in heading.
VMS/VBD match rates were calculated separately for the transit and fishing activity types.
The gear types show two different types in VBD match rates for the fishing activity type. Our
interpretation is that gear type with low match rates are not using lights to attract catch and are only
occasionally detected when nighttime operations call for extra deck lighting. This is the group with
fishing match rates less than 11%: basic longline (RD), longline tuna (RT), pole and line (H), carrier (P),
gill net (JIO/JLB), and trawlers (PI/PUD). Gear types that are routinely using lighting to attract catch,
with match rates in excess of 30% include two types of squid boats (jigging and stick dipnet), handline,
handline tuna, and two types of purse seiners. The match rates for transit activity lacks this two-typed
distribution and is consistently in the 0–10% range, comparable to the range found for gear types that
are not using lights to attract catch. Based on these results we conclude that if a vessel is detected by
VIIRS, there is a 96.42% probability that it is fishing.
The difference in light usage during fishing may also associate with the use of fish aggregating
devices (FADs). FADs are known to be a common practice, enabling fishermen to use minimal effort
for maximal catch. FADs are extensively used to attract tuna, for they tend to be attracted by floating
objects. The exact number and distribution of FADs is unclear, while licensed numbers in Indonesia
in 2006–2008 counts near 100 [
36
], some say there are 3858 or more FADs existed in Indonesian
water [
40
]. Deployment of deep-sea FADs are reported to be found in provinces like North/West
Sumatra, Lampung, east/west Java, Celebes, Maluku and Papua [
36
], which has immediate access
to deep waters as shown in Figure 1. In these waters, we found smaller numbers of VBD/VMS
matches as shown in Figure 7, and with very little to no vessel lighting observed as shown in Figure 14.
This connection implies that the two-typed distribution on match rate for purse seiners operating in
Indonesia is due to the difference in how they operate. Those operating in WPP 715, 716, and 717 are
utilizing deep-water FADs while those in WPP 712 and 713 are not. Likewise, large pelagic purse seine
with one ship (PCOB) operating in WPP 572 and 573 also more likely to rely on deep-water FADs
hence they used less lighting compared to those operating in shallow waters. Overall, as Figure 14
suggests, all bright lighting is observed in shallow waters including WPP 711, 712, and 718, while deep
water regions like WPP 572, 273, 714, 715, 716, and 717 seldom detects bright lights.
During the 32 month data period, there were 2.1 million VBD records and the match rate to VMS
vessels was 6.72%. Thus 93% of the VBD (2 million) records were not represented in the VMS record.
We believe the vast majority of the VBD lacking VMS are from vessels using lights to attract catch that
are under 30 GT level that triggers the VMS requirement. Other possible reasons why vessels may
be detected by VIIRS but lack VMS are the possibilities that the vessel has turned off their VMS or is
an illegal foreign fishing vessel. The spatial context of the VBD lacking VMS may be used to guide the
Remote Sens. 2019,11, 995 22 of 26
interpretation. For instance, VBD lacking VMS in the far northern part of the Natuna Sea are suspect
as foreign fishing vessels.
By taking the gear type match rates as a sample representing the gear types engaged in fishing
with lights it is possible to calculate the probable gear types for VBD (Table 8). Across all of Indonesia
it is most likely to find a VBD representing a small pelagic purse seiner (PCK). This is the most likely
gear type for VBD in 6 of the eleven WPP zones (571, 572, 711, 713, 715, and 716). For the remaining
five WPP the probable gear type for a VBD ranges from large pelagic purse seine with one ship (PCOB)
in WPP 573, squid dipnet (BA) in WPP 712, longline tuna (RT) in WPP 717 and 714, and squid jigging
in WPP 718.
Table 8. Dominant gear type with matched VMS/VBD in each WPP.
WPP # Region Dominant Gear Type Percentage
WPP 571 Malacca Strait Small pelagic purse seiner (PCK) 97.20%
WPP 572 West of Sumatra Small pelagic purse seiner (PCK) 48.17%
WPP 573 South of Java Large pelagic purse seine with one ship (PCOB) 78.83%
WPP 711 Natuna Sea Small pelagic purse seiner (PCK) 62.64%
WPP 712 Java Sea Stick-held squid dipnet (BA) 63.47%
WPP 713 Makassar Strait Small pelagic purse seiner (PCK) 98.80%
WPP 714 Banda Sea Longline tuna (RT) 24.43%
WPP 715 Maluku Sea Small pelagic purse seiner (PCK) 60.32%
WPP 716 Sulawesi Sea Small pelagic purse seiner (PCK) 87.14%
WPP 717 North Papua Longline tuna (RT) 50.00%
WPP 718 Arafura Sea Squid jigging (PC) 75.47%
Note: Carrier (P) is not included in this table due to being not a fishing vessel and has been banned from
further operations.
Consideration of the match rates versus average radiance indicates that most gear types have
a largely consistent style of lighting. The exception is small pelagic purse seiner (PCK), which shows
two distinct populations of vessel lighting styles in Figure 10c. There is one loose cluster with a center
of mass near 60% raw match rate and an average radiance of 50 nW/sr/cm
2
. The other population
forms a largely vertical column in Figure 10c, with raw match rates less than 40% and average radiances
less than 5 nW/sr/cm
2
. Tables 4and 5offers a geographical clue in that PCK has high match rates
with brighter lighting in the Natuna Sea and Makassar Straits and extremely low match rates in the
Maluku Sea with dimmer lighting. In tracing the geography of the vertical column cluster of Figure
10c we found these vessels are operating in the Maluku Sea. The second author conducted lighting
surveys on PCK vessels at landing sites on Java Island and in the Maluku Sea region (Ambon and
Bitung). The PCK at Java landing sites invariably carried string of bare metal halide bulbs above the
deck edges, in some cases reaching 100,000 watts in total.
Besides the using of FADs, in Ambon and Bitung the PCK still use lights to attract catch,
but deploy
small numbers of submersible lights. Field survey of lighting used on PCK vessels operating out
of Ambon and Bitung indicates the boats have 1000 to 3000 watts of deck lighting and use 1000 to
8000 watts of submersible lighting as shown in Figure 11. Our interpretation is the vertical column on
Figure 10c is an indication for the use of submersible lighting. Certainly the detection of submersible
lighting on fishing boats by satellite is quite challenging. Not only is the wattage used quite small but
also the water is opaque in the near infrared, which covers half of the DNB bandpass, which straddles
the visible and NIR. This study lays the foundation for further study to improve VBD regarding
this issue.
While no field data has been collected on PCOB, it is clear from Figure 10d that he primary cluster
is a vertical column quite similar to the one seen on Figure 10c. This suggests that PCOB vessels
are commonly using submersible lights. Another gear type with a similar vertical column cluster
is longline tuna (RT). Submersible LED lights are widely used to attract catch on RT vessels [
41
],
with on board lighting also becoming more popular in recent years as shown in Table 5. Based on
Remote Sens. 2019,11, 995 23 of 26
these new understandings, we can further investigate the impact on VBD resulted from difference
lighting methods.
5. Conclusions
A methodology has been developed for cross-matching VIIRS boat detections with
status-classified VMS tracks. The process involves predicting the probable location of a VMS equipped
vessel at the time of each VIIRS data collection. An orbital model is used to calculate VIIRS overpass
times. VMS tracks are segmented and classified by their location, speed, and heading change to
determine the status of each record. The algorithm can also be applied to AIS tracks.
Given an extended temporal record, it would be possible to develop DNB lighting profiles
for individual VMS vessels, providing additional information for more detailed analyses of vessel
behavior. Such vessel specific profiles could include the overall match rate, spatial distribution of
matches, and radiance levels. It is also possible to derive group profiles for all vessels of a specific gear
type or associated with a specific landing site.
There are several applications for cross-matching VIIRS boat detections with GPS based vessel
tracks. If the matching is done in near real time as VBD and VMS data becomes available, it will be
a great help to the authororties by identifying “dark vessels” that lack an operating VMS or AIS system.
In some areas, this is a tip-off for potentially illegal fishing. Examples may include fishery closures,
restricted waters, or Exclusive Economic Zone (EEZ) boundary zones. It is reasonable to assume that
most AIS and VMS vessels operate legally most of the time. It has been reported that vessels turn off
their AIS or VMS devices while engaged in illegal fishing. Thus a VBD lacking AIS or VMS in fishing
grounds where all vessels should reasonably have one or the other is instantly suspect in terms of IUU
fishing. Another potential application is the identification of offshore transshipment events. Several
organizations analyze AIS and VMS tracks to identify transshipment events [
42
44
]. The typical
positive identification of transshipment is two or more stationary vessels in an offshore location. The
conclusion is particularly strong if one of the vessels is registered as a carrier or transporter. But what
about the cases where an AIS or VMS vessel is “loitering” with no other AIS or VMS vessel present? If
the loitering period extends after midnight, our methods make it possible to check for VBD matches
that may indicate a transshipment event. This evidence could be interpreted relative to the DNB match
rate and radiance profile of the subject AIS or VMS vessel. Common questions from the VBD users
are what type of vessels are detected by VIIRS and can the VBD be interpreted as fishing activity, as
opposed to transit? The results from this study confirmed that in Indonesia, five gear types including
squid fishing, lift net, and purse seiners commonly deploy and operate lights to attract catch. This
includes what are the detection limits in terms of light output. The results from this study show that
given an extended period of observation, it is possible to develop match rate profiles for individual
vessels that can be used with some advantage.
The vast numbers of VBD that lack VMS testify to the value of the VBD data in monitoring
fishing grounds where lighting is used to attract catch. In fishing grounds where lights are used to
attract catch, VBD provide the most complete near real time data source for locating fishing activity.
EOG posts the latest VBD output within 4 to 6 h from satellite overpass. The time includes data
gathering and processing.
Further development of this study can be used to estimate unreported catch and improve stock
assessments as well as other related information relevant for decision makers. Moreover, improved
monitoring programs can benefit the assessment of efficient harvest strategies through improved
estimates of maximum sustainable yield (MSY), biomass and carrying capacity of the fishery [
45
];
while the problems with catch under-reporting appear to be particularly serious in Indonesia, especially
for catches of tuna and tuna like species [46].
Verifying the reported logbooks against landing data using VBD and VMS data proposed by this
research, could address the problem of uncertainty in the level of reporting from fishers to the fishing
port authority. The urgency of such verification using VBD and VMS data is due to the current trip
Remote Sens. 2019,11, 995 24 of 26
length which was estimated by the number of fishing days by departure and arrival dates of vessels,
while some purse-seine vessels, however, transfer their catch to a carrier vessel that brings the catch
ashore without any proper catch reporting to the authority.
Author Contributions:
F.C.H. developed the cross matching and VMS track segmentation software, processed
the VMS records, and was the primary writer of the paper. C.E. wrote the introduction, discussion and conclusion.
M.Z. wrote the VBD algorithms. K.B. processed the VIIRS data to generate the VBD. T.G. assisted K.B. in generating
the VBD data. D.K. provided technical details on the VMS data. A.S.,W.B., and M.R. reviewed the manuscript and
provided local knowledge. R.N. and Y.S. helped provide local knowledge and VMS data.
Acknowledgments:
This research was funded in part by the NOAA JPSS proving ground program and the U.S.
Agency for International Development. The research was possible thanks to the collection of VMS data by the
Indonesia Ministry of Marine Affairs and Fisheries. Our access to the VMS records was arranged for use in
research through Global Fishing Watch. The paper benefited from a detailed review provided by John Mittleman,
Naval Research Laboratory, Washington, DC.
Conflicts of Interest: The authors declare no conflict of interest.
Abbreviations
The following abbreviations are used in this manuscript:
AIS Automatic identification system
DMSP-OLS Defense Meteorological Satellite Program-Operational Linescan System
DNB Day night band
FAD Fishg aggregating device
GFW Global Fishing Watch
GT Gross tonnage
EEZ Exclusive Economic Zone
EOG Earth Observation Group
IMO International Maritime Organization
IUU Illegal, Unreported and Unregulated
MMAF Ministry of Maritime Affairs and Fisheries
MSY Maximum sustainable yield
NASA National Aeronautics and Space Administration
NOAA National Oceanic and Atmospheric Administration
SAR Synthetic aperture radar
Suomi NPP Suomi National Polor-orbiting Partnership
TLE Two line element
UTM Universal transverse mercator
VBD VIIRS boat detection
VIIRS Visual infrared imaging radiometer suite
VMS Vessel monitoring system
WPP Wilayah Pengelolaan Perikanan (Fishery Management Area)
References
1.
Suhendar, M. Comparison of Vessel Monitoring System (VMS) between Iceland and Indonesia [Final
Project]. Available online: http://www.unuftp.is/static/fellows/document/suhendar12prf.pdf (accessed
on 15 August 2018).
2.
Enforcement | NOAA Fisheries. Available online: https://www.fisheries.noaa.gov/topic/enforcement#
vessel-monitoring (accessed on 21 August 2018).
3.
Automatic Identification System (AIS). Available online: http://www.imo.org/en/OurWork/safety/
navigation/pages/ais.aspx (accessed on 10 August 2018).
4.
Tello, M.; Lopez-Martinez, C.; Mallorqui, J.J. A novel algorithm for ship detection in SAR imagery based on
the wavelet transform. IEEE Geosci. Remote Sens. Lett. 2005,2, 201-205. [CrossRef]
5.
Greidanus, H.; Clayton P.; Indregard M.; Staples G.; Suzuki N.; Vachoir P.; Wackerman C.; Tennvassas T.;
Mallorqui J.; Kourti N.; et al. Benchmarking operational SAR ship detection. In Proceedings of the IEEE
International Geoscience and Remote Sensing Symposium, Anchorage, AK, USA, 20–24 September 2004.
Remote Sens. 2019,11, 995 25 of 26
6.
Wu, G.; de Leeuw, J.; Skidmore, A.K.; Liu, Y.; Prins, H.H.T. Performance of LandsatTM in ship detection in
turbid waters. Int. J. Appl. Earth Obs. Geoinf. 2009,11, 54–61. [CrossRef]
7.
Máttyus, G. Near real-time automatic marine vessel detection on optical satellite images. In Proceedings of
the International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, ISPRS
Hannover Workshop, Hannover, Germany, 21–24 May 2013.
8.
Corbane, C.; Najman, L.; Pecoul, E.; Demagistri, L.; Petit, M. A complete processing chain for ship detection
using optical satellite imagery. Int. J. Remote Sens. 2010,31, 5837–5854. [CrossRef]
9.
Zhu, C.; Zhou, H.; Wang, R.; Guo, J. A novel hierarchical method of ship detection from spaceborne optical
image based on shape and texture features. IEEE Trans. Geosci. Remote Sens.
2010
,48, 3446–3456. [CrossRef]
10. Croft, T.A. Night-time images of the earth from space. Sci. Am. 1978,239, 68–79. [CrossRef]
11.
Elvidge, C.D.; Baugh, K.; Zhizhin, M.; Hsu, F.-C. Why VIIRS data are superior to DMSP for mapping
nighttime lights. Proc. Asia-Pac. Adv. Netw. 2012,35, 62–69. [CrossRef]
12.
Oozeki, Y.; Inagake, D.; Saito, T.; Okazaki, M.; Fusejima, I.; Hotai, M.; Watanabe, T.; Sugisaki, H.; Miyahara, M.
Reliable estimation of IUU fishing catch amounts in the northwestern Pacific adjacent to the Japanese EEZ:
Potential for usage of satellite remote sensing images. Mar. Policy 2018,88, 64–74. [CrossRef]
13.
Guo, G.; Fan, W.; Xue, J.; Zhang, S.; Zhang, H.; Tang, F.; Cheng, T. Identification for operating pelagic
light-fishing vessels based on NPP/VIIRS low light imaging data. Trans. Chin. Soc. Agric. Eng.
2017
,33,
245–251.
14.
Cozzolino, E.; Lasta, C.A. Use of VIIRS DNB satellite images to detect jigger ships involved in the Illex
argentinus fishery. Remote Sens. Appl. Soc. Environ. 2016,4, 167–178. [CrossRef]
15.
Liu, Y.; Saitoh, S.I.; Hirawake, T. Detection of squid and pacific saury fishing vessels around Japan using
VIIRS Day/Night Band image. Proc. Asia-Pac. Adv. Netw. 2015,39, 28–39. [CrossRef]
16.
Elvidge, C.; Zhizhin, M.; Baugh, K.; Hsu, F-C. Automatic Boat Identification System for VIIRS Low Light
Imaging Data. Remote Sens. 2015,7, 3020–3036. [CrossRef]
17.
Joint Polar Satellite System. Available online: http://www.jpss.noaa.gov/launch.html (accessed on
21 August 2018).
18.
EOG VIIRS Boat Detection. Available online: https://eogdata.mines.edu/vbd/ (accessed on
4 February 2019
).
19.
Guerriero, M.; Wilett, P.; Coraluppi, S.; Carthel, C. Radar/AIS data fusion and SAR tasking for Maritime
Surveillance. In Proceedings of the International Conference on Information Fusion, Cologne, Germany,
30 June–3 July 2008.
20.
Grasso, R.; Mirra, S.; Baldacci, A.; Horstmann, J.; Coffin, M.; Jarvis, M. Performance Assessment of
a Mathematical Morphology Ship Detection Algorithm for SAR Images through Comparison with AIS Data.
In Proceedings of the International Conference on Intelligent Systems Design and Applications, Pisa, Italy,
30 November–2 December 2009.
21.
Margarit, G.; Tabasco, A. Ship Classification in Single-Pol SAR Images Based on Fuzzy Logic. IEEE Trans.
Geosci. Remote Sens. 2011,49, 3129–3138. [CrossRef]
22.
Chaturvedi, S.K.; Yang, C.-S.; Ouchi, K.; Shanmugam P. Ship Recognition by Integration of SAR and AIS.
J. Navig. 2012,65, 323–337. [CrossRef]
23.
Lemoine, G.; Ghesworth, J.; Schwartz-Juste, G.; Kourti, N.; Shepherd, I. Near real time vessel detection using
spaceborne SAR imagery in support of fisheries monitoring and control operations. In Proceedings of the
IEEE International Geoscience and Remote Sensing Symposium, Anchorage, AK, USA, 20–24 Septemer 2004.
24.
KKP WPP MapServer. Available online: http://www.ppk-kp3k.kkp.go.id/ArcGIS/rest/services/kkp/
Kawasan/MapServer (accessed on 26 March 2019).
25.
Elvidge, C.; Zhizhin, M.; Baugh, K.; Hsu, F.C.; Ghosh, T. Methods for global survey of natural gas flaring
from visible infrared imaging radiometer suite data. Energies 2016,9, 14. [CrossRef]
26.
Vallado, D.A.; Crawford, P.; Hujsak, R.; Kelso, T.S. Revisiting Spacetrack Report #3. In Proceedings of the
AIAA/AAS Astrodynamics Specialist Conference, Keystone, CO, USA, 21–24 August 2006.
27. Kelso, T.S. Frequently Asked Questions: Two-Line Element Set Format. Satellite Times 1997,4, 68–69.
28. CelesTrak. Available online: http://celestrak.com/ (accessed on 5 April 2015).
29.
McCluskey, S.M.; Lewison, R.L. Quantifying fishing effort: A synthesis of current methods and their
applications. Fish Fish. 2008,9, 188–200. [CrossRef]
30.
De Souza, E.N.; Boerder, K.; Matwin, S.; Worm, B. Improving Fishing Pattern Detection from Satellite AIS
Using Data Mining and Machine Learning. PLoS ONE 2016,11, e0158248.
Remote Sens. 2019,11, 995 26 of 26
31.
Calenge, C. Analysis of Animal Movements in R: The AdehabitatLT Package. Available online: https://cran.
r-project.org/web/packages/adehabitatLT/vignettes/adehabitatLT.pdf (accessed on 31 August 2018).
32.
NASA Ocean Color. Available online: https://oceancolor.gsfc.nasa.gov/docs/distfromcoast/ (accessed on
17 June 2017).
33.
Pusat Informasi Pelabuhan Perikanan. Available online: http://pipp.djpt.kkp.go.id (accessed on
15 August 2017).
34. OpenStreetMap. Available online: https://www.openstreetmap.org/ (accessed on 15 May 2018).
35.
Baker, N.; Kilcoyne H. Joint Polar Satellite System (JPSS) VIIRS Radiometric Calibration Algorithm Theoritical
Basis Document (ATBD); Rev. C; Goddard Space Flight Center: Greenbelt, MD, USA, 2013.
36.
Yasfiandayani, R.; Baskoro, M.; Monintja, D. Imapct of fish aggregating device on sustainable capture
fisheries. In Proceedings of the 1st International Symposium on Aquatic Product Processing 2013, Bogor,
Indonesia, 13–14 November 2013. [CrossRef]
37.
Natsir, M.; Atmaja, S.B. Fishing activity of individual purse seiners in Molucca Sea: Vessel Monitoring
System (VMS) and observers. J. Penelitian Perikanan Indonesia 2013,19, 17–24.
38.
Mulyadi, R.A.; Brown, A.; Rengi, P. Study Technology Hand Line In Ocean Fishing Port Bungus Province
West Sumatra. J. Online Mahasiswa Fakultas Perikanan Ilmu Kelautan Univ. Riau 2015,2, 1–13.
39.
Mulyadi, R.A.; Sondita, M.F.A.; Yusfiandayani, R. Suhu permukaan laut dan kedalaman tertangkapnya tuna
oleh kapal pancing ulur yang dilengkapi lampu. J. Teknol. Perikanan Kelautan 2018,8, 199–207. [CrossRef]
40.
Fish Magnet Boom Creates Headaches in Indonesia’S War on Overfishing. Available online: https://
news.mongabay.com/2017/02/fish-magnet-boom-creates-headaches-in-indonesias-war-on- overfishing/
(accessed on 24 February 2019)
41.
Submersible Underwater Green Fishing Lights Longline Tuna Lamp For Deep Sea. Available
online: http://underwaterledfishinglights.sell.everychina.com/p-106898950-submersible-underwater-
green-%20\fishing-lights-longline-tuna-lamp-for-deep-sea.html (accessed on 12 September 2018).
42.
Miller, N.A.; Roan, A.; Hochberg, T.; Amos, J.; Kroodsma D. A. Identifying Global Patterns of Transshipment
Behavior, Front. Mar. Sci. 2018,5, 240.
43.
Boerder, K.; Miller, N.A.; Worm, B.; Global hot sopts of transshipment of fish catch at sea. Sci. Adv.
2018
,
4, 7159. [CrossRef]
44.
European Fisheries Control Agency. Core Curriculum for the Training of Fisheries Inspectors: 2b Transhipment
Inspection; Europena Union: Luxembourg, 2017; ISBN 978-92-9209-944-9
45.
Rudd, M.B.; Branch, T.A. Does unreported catch lead to overfishing? Fish Fish.
2017
,18, 313-323. [CrossRef]
46.
Yuniarta, S.; van Zwieten, P.A.; Groeneveld, R.A.; Wisudo, S.H.; van Ierland, E.C. Uncertainty in catch
and effort data of small- and medium-scale tuna fisheries in Indonesia: Sources, operational causes and
magnitude. Fish. Res. 2017,193, 173–193. [CrossRef]
©
2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).
... Studies related to the use of Big Data in the field of fisheries include: prediction of the crab season for cancer magister [6], big data management to help fisheries in China [7]. Broader big data management that covers all oceans in the European Union and several parts of the world has been carried out, where big data management helps determine shipping lanes, detect weather and water anomalies, maritime conservation [8], and map marine biota [9]. ...
... Several technologies owned by KKP have been used to assist the development of the fishing industry, namely the visible infrared imaging radiometer suite (VIIRS) and boat detections (VBD) with a vessel monitoring system (VMS). The VIIRS system is able to detect illegal fishing across the 11 Indonesian Fisheries Management Areas (WPPNRI) [9] as depicted in Figure 2. ...
... Distribution Patterns of Certain Types of Fish and Recommended Fishing Techniques to Fishermen[9]. ...
... Thus, there are still many gaps in the monitoring of ships at daily, monthly and annual time scales. Recent studies have promoted the use of VIIRS low light imaging data for monitoring ships that are using artificial lights within the fisheries industry [19,20]. In this study, we propose the use of night light data to monitor shipping activity in anchorage areas, thereby filling the research gap with a new method for assessing statistical parameters of shipping activities at the port level. ...
Article
Full-text available
Accurate information on port shipping activities is critical for monitoring global and local traffic flows and assessing the state of development of the maritime industry. Such information is necessary for managers and analysts to make strategic decisions and monitor the maritime industry in achieving management goals. In this study, we used monthly night light (NTL) images of the Suomi National Polar-Orbiting Partnership (Suomi NPP) Visible Infrared Imaging Radiometer Suite (VIIRS) Day/Night Band, between 2012 and 2020, to study the night lights emitted by ships in ports' anchorage areas, as an indicator for shipping activity in anchorage areas and ports. Using a dataset covering 601 anchorage areas from 97 countries, we found a strong correspondence between NTL data and shipping metrics at the country level (n = 97), such as container port throughput (Rs = 0.84, p < 0.01) and maximum cargo carried by ships (Rs = 0.66, p < 0.01), as well as a strong correlation between the number of anchorage points and the NTL values in anchorage areas across the world (Rs = 0.69, p < 0.01; n = 601). The high correspondence levels of the VIIRS NTL data with various shipping indicators show the potential of using NTL data to analyze the spatio-temporal dynamic changes of the shipping activity in anchorage areas, providing convenient open access and a normalized assessment method for shipping industry parameters that are often lacking.
... VIIRS/DNB data have been increasingly utilized in various fishery studies since the launch of the Suomi National Polar Orbiting Partnership satellite in 2011. Its applications include detecting changes in the number and positions of fishing vessels, identifying illegal, unreported and unregulated fishing, tracking the operation status of saury fishing vessels, mapping fishing activities and suitable fishing grounds, and tracking fishing vessel distributions in combination with vessel monitoring system data (Liu et al., 2015;Straka et al., 2015;Cozzolino and Lasta, 2016;Syarifudin et al., 2017;Geronimo et al., 2018;Tian et al., 2018;Hsu et al., 2019;Saito et al., 2019). Threshold segmentation and DNB local maximum extraction are key steps of the VIIRS/DNB boat detection (VBD) algorithm developed by Elvidge et al. (2015) to extract offshore night light fishing boats. ...
Article
Full-text available
Monitoring the dynamics of ocean-going fishing vessels is essential for fisheries stock assessment and management. In this paper, coupled fishery and remote sensing methods were applied to propose a satellite-based fishery resource monitoring and assessment system (SFRMAS) that supported chub mackerel (Scomber japonicus), Pacific saury (Cololabis saira) and neon flying squid (Ommastrephes bartramii) fishing activities in the northwest Pacific. A multispecies fishing vessel database was generated in the SFRMAS using visible infrared imaging radiometer suite day/night band (VIIRS/DNB) images, automatic identification system (AIS) data and multimarine remote sensing resources. The results showed that the VIIRS/DNB vessel detection (VVD) and AIS vessel detection approaches improved the extraction accuracy of multispecies fishing vessels. A high productivity area with a higher chlorophyll a concentration had the most concentrated distribution of mackerel vessels. The widest longitudinal migrations of the neon flying squid and the latitudinal ranges of Pacific saury were highly related to sea surface temperature (SST), which was well captured by the movement of the fishing vessels. In the SFRMAS, the accuracy of detected multispecies fishing days and fishing catches was mostly between 60 and 90%, which can provide data support for fisheries stock assessment and management.
... Across the social and environmental sciences, studies by economists, political scientists, geographers, and others have used satellite imagery to expose the presence and/or effects of corruption (Hodler and Raschky, 2014), inequality (Ivan et al., 2020;Zhou et al., 2015), poverty (Briggs, 2018), unequal distribution of public services like electricity (Min et al., 2013;Min and Gaba, 2014), natural disasters (Gillespie et al., 2014), conflict (Agnew et al., 2008Cederman et al., 2015;Coscieme et al., 2017), refugee movements, settlements, and forced migrations (Ahmed et al., 2019;Van Den Hoek and Friedrich, 2021;Witmer and O'Loughlin, 2011), and mass atrocities and genocide (Marx et al., 2019). Remote sensing has also been used to expose deforestation driven by refugee camps (Ahmed et al., 2019;Hassan et al., 2018), illegal mining (Kyba et al., 2019), gas flaring (Elvidge et al., 2013), and illegal fishing (Hsu et al., 2019;Lumban-Gaol et al., 2020). While much of this research is not informed by critical theory, its attention to issues of social and environmental justice overlaps with critical human geographers' commitment to progressive social change. ...
Article
We offer a review and research agenda for critical remote sensing, defined as inquiries and scientific practices cognizant of the embedding of power within the production, analysis, and instrumentalization of satellite imagery. First, we consider critiques of the satellite gaze. Second, we chronicle remote sensing’s evolving political economy, examining the technology’s use by governments, scientists, and commercial and non-governmental actors. Then, we review practices of critical remote sensing, categorized as research 1) exposing injustices; 2) engaging situated knowledges; and 3) empowering marginalized actors. Lastly, we suggest five areas for intertwining critiques and practices and consider possibilities for counter remote sensing.
... The role of satellite-based monitoring technology is very practical in detecting the occurrence of illegal fishing. Several technologies have been used widely for monitoring vessel activities, such as vessel monitoring system (VMS), automatic identification system (AIS), Radar-SAR imagery, very high-resolution optical imagery, and Visible Infrared Imaging Radiometer Suite (VIIRS) [4,5,6]. VMS and AIS are GPS-based monitoring technologies [7]. ...
Conference Paper
Full-text available
Detecting fishing boat activity is still a challenge for the biggest archipelago countries, such as Indonesia, to monitor the huge marine area. Space technology using sensors SAR to detect ships has been developed since 1985. However, the cost of using SAR images is one of the barriers to operational aspects, mainly for detecting fishing boats to deter IUU fishing activities. This research aims to evaluate the use of Sentinel 1-SAR imagery for identifying fishing boats from space. We used VIIRS data for validating the purposes. Both data sources could be accessed freely. The object detection process can be derived into three steps: pre-processing, object detection and object validation. We used the constant false alarm rate (CFAR) method to discriminate against objects at sea. To identify fishing vessels, we used the size of the vessels and the intensity of light captured by VIIRS. According to the findings, 21 boats were discovered using sentinel 1-SAR imagery and four boats using VIIRS data based on the the area of interest.
Article
The seasonal variations in shelf circulation and their implications for the distribution of light fishing vessels positions (VBD) that capture small pelagic fishes in the southern South China Sea (SSCS) are investigated by using multi-datasets between 2008–2014. It is found that the SSCS shelf circulation is indicated by the strong seasonal reversal western boundary current (WBC) and prominent cyclonic circulation (eddy). The southeastward WBC brings homogeneous salty, colder, and high primary production of nanophytoplankton (PPN) water during winter-time from the northern and the cyclonic eddy region into the southern SSCS. In contrast, less salty, warmer, and less PPN are advected northwestward from Karimata into the northern region. These factors significantly influence the distribution of VBD and in turn the production of captured small pelagic fishes (CPUE), revealing two maxima (peaks) of both VBD and CPUE during the monsoon breaks (MB) with a time-lag of 2–3 month from the monsoon peaks. Relationship between oceanographic parameters and CPUE is examined statistically revealing that ocean current and PPN are significantly affect the CPUE. A relatively high PPN and strong current during the peaks of monsoon are correlated with lower CPUE. In contrast, during the MBs period CPUE maxima are associated with a relatively weak ocean current and consumed PPN that favor the abundance of small pelagic fishes.
Preprint
Full-text available
Host to intricated networks of marine species, coral reefs are among the most biologically diverse ecosystems on Earth. Over the past decades, major degradations of coral reefs have been observed worldwide, which is largely attributed to the effects of climate change and local stressors related to human activities. Now more than ever, characterizing how the environment shapes the dynamics of the reef ecosystem is key to (1) uncovering the environmental drivers of reef degradation, and (2) enforcing efficient conservation strategies in response. To achieve these objectives, it is pivotal that environmental data characterizing such ecosystem dynamics, which occur across specific spatial and temporal scales, are easily accessible to coral reef researchers and conservation stakeholders alike. Here we present the Reef Environment Centralized Information System (RECIFS), an online repository of datasets describing reef environments worldwide over the past few decades. The data served through RECIFS originate from remote sensed datasets available in the public domain, and characterize various facets of the reef environment, including water chemistry and physics (e.g. temperature, pH, chlorophyll concentration), as well as anthropogenic local pressures (e.g. boat detection, distance from agricultural or urban areas). The datasets from RECIFS can be accessed at different spatial and temporal resolutions and are delivered through an intuitive web-application featuring an interactive map requiring no prior knowledge working with remote sensing or geographic information systems. The RECIFS web-application is available in complete open access at https://recifs.epfl.ch. We describe two case studies showing possible implementations of RECIFS to 1) characterize coral diversity in the Caribbean and 2) investigate local adaptation of a reef fish population in Northwest Australia.
Article
Full-text available
Tuna and tuna-like fisheries play a vital role in Indonesian livelihood, especially in the archipelagic waters. However, despite the importance, the concern in general data collection activities for tuna, i.e., limited, with incomplete scientific knowledge and insufficient data has hampered the assessment. The purpose of this study was to analyse on how fisheries-dependent data system could transform the data quality. For this purpose, the analysis was conducted based on e-logbook data from January to December 2020. The result showed that some robust fisheries indicators (catch, effort, fleet and species composition) could be produced from this particular dataset. In addition, e-logbook has the best attribute for improving the data quality, especially for small-scale tuna fisheries in FMA-714. Low cost and vast spatial and temporal coverage are the main characteristics on why the program should be expanded and monitored carefully in the future.
Article
With the progression of overfishing along the coast, oceanic fisheries in the South China Sea have attracted increasing attention from bordering countries. Fishing with lights has developed rapidly over the past decade. In this study, we analysed the trend in fishing over the spring fishing season (March-April) from  to  in the open-SCS fishing zone based on nightly satellite. The results indicated that the number of fishing boats detected by satellites were apparently influenced by the phases of the moon. Using data from moonless nights, we estimated that the number of fishing boats increased from ∼ to ∼ over the past decade. These fishing craft in the open-SCS could be classified into large falling net vessels with bright lights and tuna fishing boats with dim lights. The nightly images of large falling net vessels were studied further using records from a typical commercial fishing vessel and, on this basis, we established an algorithm to extract data for this type of craft, whose numbers were estimated to have increased from ∼ to ∼ over the past decade. Using this algorithm, we were able to trace the development of these fleets and map out their distribution patterns in the open-SCS.
Article
Full-text available
p>Sebaran dan kelimpahan ikan tuna sangat dipengaruhi oleh suhu dan kedalaman air. Nelayan tuna Bungus menangkap ikan tuna bigeye (BET, Thunnus obesus) dan tuna sirip kuning (YFT, Thunnus albacares) menggunakan pancing ulur yang dioperasikan pada kapal yang dilengkapi lampu sebagai alat abantu. Tujuan penelitian: (1) mengukur suhu permukaan laut (SST) dan kedalaman renang tuna mata besar dan tuna sirip kuning di sekitar Pulau Mentawai, (2) mengidentifikasi pengaruh cahaya terhadap sebaran vertikal tuna, dan (3) menentukan panjang tali pancing terbaik untuk penangkapan tuna. Berdasarkan panjang garis dan konfigurasi garis, berat, kait dan umpan cumi-cumi, kait mereka berada di 45, 53, 60, dan 68 meter di bawah permukaan laut. Sebanyak 8 tuna sirip kuning tertangkap di kedalaman 45, 53 dan 60 m; berat total 354 kg. Satu BET seberat 45 kg tertangkap pada kedalaman 60 m. Penelitian ini memberitahukan bahwa tuna berukuran besar tertangkap di lapisan permukaan dengan kedalaman 15-60 m. Suhu permukaan laut (SST) di daerah penangkapan ikan di sekitar Pulau Mentawai rata-rata 28.97 OC di mana nelayan berhasil menangkap 15 tuna yang terdiri dari 3 ekor tuna mata besar dan 12 ekor tuna sirip kuning. Tuna dewasa kebanyakan tertangkap pada kedalaman 23-60 m sedangkan muda tertangkap di kedalaman 15-45 m. Penelitian ini menunjukkan pengaruh cahaya dapat menaikkan posisi lapisan renang tuna dewasa. Handline dengan panjang tali 53 m adalah panjang tali terbaik untuk menangkap tuna dewasa di daerah tersebut.</p
Article
Full-text available
A major challenge in global fisheries is posed by transshipment of catch at sea from fishing vessels to refrigerated cargo vessels, which can obscure the origin of the catch and mask illicit practices. Transshipment remains poorly quantified at a global scale, as much of it is thought to occur outside of national waters. We used Automatic Identification System (AIS) vessel tracking data to quantify spatial patterns of transshipment for major fisheries and gear types. From 2012 to 2017, we observed 10,510 likely transshipment events, with trawlers (53%) and longliners (21%) involved in a majority of cases. Trawlers tended to transship in national waters, whereas longliners did so predominantly on the high seas. Spatial hot spots were seen off the coasts of Russia and West Africa, in the South Indian Ocean, and in the equatorial Pacific Ocean. Our study highlights novel ways to trace seafood supply chains and identifies priority areas for improved trade regulation and fisheries management at the global scale.
Article
Full-text available
Transshipment at sea, the offloading of catch from a fishing vessel to a refrigerated vessel far from port, can obscure the actual source of the catch, complicating sustainable fisheries management, and may allow illegally caught fish to enter the legitimate seafood market. Transshipment activities often occur in regions of unclear jurisdiction where policymakers or enforcement agencies may be slow to act against a challenge they cannot see. To address this limitation, we processed 32 billion Automatic Identification System (AIS) messages from ocean-going vessels from 2012 to the end of 2017 and identified and tracked 694 cargo vessels capable of transshipping at sea and transporting fish (referred to as transshipment vessels). We mapped 46,570 instances where these vessels loitered at sea long enough to receive a transshipment and 10,233 instances where we see a fishing vessel near a loitering transshipment vessel long enough to engage in transshipment. We found transshipment behaviors associated with regions and flag states exhibiting limited oversight; roughly 47% of the events occur on the high seas and 42% involve vessels flying flags of convenience. Transshipment behavior in the high seas is relatively common, with vessels responsible for 40% of the fishing in the high seas having at least one encounter with a transshipment vessel in this time period. Our analysis reveals that addressing the sustainability and human rights challenges (slavery, trafficking, bonded labor) associated with transshipment at sea will require a global perspective and transnational cooperation.
Article
To establish an estimation procedure for reliable catch amount of illegal, unreported and unregulated (IUU) fishing, light-gathering fishing operations in the northwestern Pacific were analyzed based on the Visible Infrared Imaging Radiometer Suite (VIIRS) day/night band (DNB) data provided by the Suomi National Polar Partnership (SNPP) satellite. The estimated fishing activities were compared with the navigation tracks of vessels obtained from the automatic identification system (AIS). As a model case, the fishing activities of Chinese fishing boats using fish aggregation lights outside the Japanese EEZ in the northwestern Pacific were analyzed from mid-June to early-September 2016. Integration analyses of VIIRS DNB data and AIS information provided reliable data for estimating the fishing activities of Chinese fishing boats and suggested the importance of estimating fish carrier ship movements. The total amount of the chub mackerel (Scomber japonicus) catch during this period was independently estimated from three angles: 1) the fishing capacity of the fishing boats, 2) the freezing capacity of refrigeration factory ships and 3) the fish hold capacity of the fish carrier ships, based on information obtained from interviews with Chinese fisheries companies. These estimates indicated that the total amount of mackerel catch by Chinese fisheries was more than 80% of the allowable biological catch (ABC) of Japan in this area in 2016. This suggests that Pacific high seas fishing has a significant impact on the future of fish abundance. Our proposed procedure raises the possibility of evaluating the fishing impact of some forms of IUU fisheries independently from conventional statistical reports.
Article
Fishing data are the basement of fisheries science research, but currently the source of fishing data is extraordinarily scarce, and data quality is poor in some aspects. Satellite low light sensors can detect the light-fishing vessels at night, however, its application in pelagic fishery has been limited by the lack of an algorithm for extracting the location and brightness of operating pelagic light-fishing vessels. An examination of operating pelagic light-fishing vessels features in the day/night band (DNB) image, which was from the visible infrared imaging radiometer suite (VIIRS) on the Suomi National Polar-orbiting Partnership (NPP) satellite, indicated that the features were a list of nonadjacent bright spots. In order to identify the operating pelagic light-fishing vessels from VIIRS/DNB accurately, we designed a set of identification algorithm for operating pelagic light-fishing vessels according to the light radiation characteristics of its fishing gathering lamps in NPP/VIIRS low light image. Before applying the identification algorithm, a data pre-processing step was adopted through radiation stretch and noise reduction by adaptive Wiener filter to prepare the data for further analysis and use. A spike median index (SMI) was used to enlarge the radiation difference between operating pelagic light-fishing vessel pixels and background pixels. On the basis of this, an adaptive threshold segmentation method called the maximum entropy (MaxEnt) method was used to extract the bright spot pixels, and generated a list of candidate operating pelagic light-fishing vessels detections. The candidate pixels were then filtered to remove the false identification bright spot pixels distributed near the operating pelagic light-fishing vessel pixels, and illuminated by the high-power fishing gathering lamps by a local spike detection (LSD) algorithm. A validation study was conducted at a night with weak lunar illuminance on May 24, 2015 which was selected randomly, using the vessel monitoring system (VMS) data of Chinese operating light-seiners vessels on the high seas of Northwest Pacific Ocean light seine fishing ground and the result of VIIRS/DNB image visual interpretation. The validation result showed that the identification algorithm detected 27 operating pelagic light-fishing vessels on the high seas of Northwest Pacific Ocean light seine fishing ground, and the number of operating pelagic light-fishing boats and their distribution were entirely consistent with the result of VIIRS/DNB image visual interpretation; the VMS data had the record of 25 operating pelagic light-fishing vessels among the total 27 vessels, and their distribution was nearly the same with the result of identification algorithm and VIIRS/DNB image visual interpretation. The identification algorithm worked well when lunar illuminance was weak and its identification accuracy was above 92%. The identification algorithm not only avoided the subjectivity and uncertainty of certain threshold segmentation, but also removed the false identification bright spot pixels near the operating pelagic light-fishing vessel pixels, which were illuminated by the high-power fishing gathering lamps. Detection of operating pelagic light-fishing vessels based on VIIRS/DNB imaging data can provide up-to-date activity and change information of operating pelagic light-fishing vessels for pelagic light-fishing industry, which meets the need of fishing boat's daily monitoring, and has a wide application prospect in fishing effort estimation, research of central fishing ground spatial-temporal distribution and change, and fishery forecast and management. © 2017, Editorial Department of the Transactions of the Chinese Society of Agricultural Engineering. All right reserved.
Article
This study aims to identify the sources and magnitude of uncertainty in the collection and processing of catch and effort data of small- and medium-scale tuna fisheries in Indonesia, as well as the causes of uncertainty on an operational level. We identified possible sources of uncertainty through a literature review and interviews with experts. Next, we surveyed 40 small-scale (<10 GT) and medium-scale (10–100 GT) pole-and-line, purse-seine, longline and handline fishers in the oceanic fishing port Bitung, which has the largest number of tuna fisheries activities in eastern Indonesia, to estimate the magnitude of unreported catch of juvenile tuna, on-board consumption, home consumption and catch used as bait. We used logbook data from the fisheries submitted to the fishing port authorities to extrapolate survey results to the fishing port level. Uncertainties around unreported catches were due both to non-reporting by fishers to the fishing port authority and to flaws in data management in the data collection institution. After removing flaws in the logbook database we estimated that the catch by small- and medium-scale fishing vessels active in Indonesian waters could be about 33–38% higher than reported. The proportion of unreported catch, as well as the sources and range of uncertainty, varied according to the types of gear used. Finally, we discuss what aspects of data collection and processing should be improved at the fishing port level, including the identified sources of unreported catch and the processes leading to non-reporting. We hence provide a methodology for estimating unreported catches in small and medium-scale fisheries.
Conference Paper
Vessel monitoring and surveillance is important for maritime safety and security, environment protection and border control. Ship monitoring systems based on Synthetic-aperture Radar (SAR) satellite images are operational. On SAR images the ships made of metal with sharp edges appear as bright dots and edges, therefore they can be well distinguished from the water. Since the radar is independent from the sun light and can acquire images also by cloudy weather and rain, it provides a reliable service. Vessel detection from spaceborne optical images (VDSOI) can extend the SAR based systems by providing more frequent revisit times and overcoming some drawbacks of the SAR images (e.g. lower spatial resolution, difficult human interpretation). Optical satellite images (OSI) can have a higher spatial resolution thus enabling the detection of smaller vessels and enhancing the vessel type classification. The human interpretation of an optical image is also easier than as of SAR image. In this paper I present a rapid automatic vessel detection method which uses pattern recognition methods, originally developed in the computer vision field. In the first step I train a binary classifier from image samples of vessels and background. The classifier uses simple features which can be calculated very fast. For the detection the classifier is slided along the image in various directions and scales. The detector has a cascade structure which rejects most of the background in the early stages which leads to faster execution. The detections are grouped together to avoid multiple detections. Finally the position, size(i.e. length and width) and heading of the vessels is extracted from the contours of the vessel. The presented method is parallelized, thus it runs fast (in minutes for 16000x16000 pixels image) on a multicore computer, enabling near real-time applications, e.g. one hour from image acquisition to end user.
Article
The distribution of Illex argentinus squid extends from 23°S to 54°S. The largest catches of the species, which represents one of the most important fisheries in Argentina, take place between 35°S and 52°S. Argentina's fisheries administration keeps close records of the Argentine fleet position and the Cephalopod laboratory at the Instituto Nacional de Investigación y Desarrollo Pesquero (INIDEP) monitors and suggests actions for the management of the resource. The catches are carried out both within national and adjacent international waters. Fleets from different countries participate in the fisheries operating jigger vessels during the night with strong lights to attract the squid. One of the greatest difficulties in the evaluation of the status of this resource is to know the number of foreign vessels fishing outside the Argentine Exclusive Economic Zone (EEZ). The Visible Infrared Imaging Radiometer (VIIRS) day/night band (DNB) satellite images are a useful tool to monitor and quantify these fleets, building on the capacity of the sensors to detect the light emitted by the lamps placed on the ship decks. In this work, we report the development of a specific new method (set of algorithms) to process the images and identify automatically the jigger ships that compose the overseas fleet. Results were validated using the positioning data of the Argentine jigger fleet and comparing light emissions of these vessels against those identified by the new method. The process of identifying ships has proved to be robust considering the statistical results obtained: mean relative error (MRE) of 0.03% and a root-mean-square error (RMSE) of 1.62 ships.
Article
Catches are commonly misreported in many fisheries worldwide, resulting in inaccurate data that hinder our ability to assess population status and manage fisheries sustainably. Under-reported catch is generally perceived to lead to overfishing, and hence, catch reconstructions are increasingly used to account for sectors that may be unreliably reported, including illegal harvest, recreational and subsistence fisheries, and discards. However, improved monitoring and/or catch reconstructions only aid in the first step of a fisheries management plan: collecting data to make inferences on stock status. Misreported catch impacts estimates of population parameters, which in turn influences management decisions, but the pattern and degree of these impacts are not necessarily intuitive. We conducted a simulation study to test the effect of different patterns of catch misreporting on estimated fishery status and recommended catches. If, for example, 50% of all fishery catches are consistently unreported, estimates of population size and sustainable yield will be 50% lower, but estimates of current exploitation rate and fishery status will be unbiased. As a result, constant under-or over-reporting of catches results in recommended catches that are sustainable. However, when there are trends in catch reporting over time, the estimates of important parameters are inaccurate, generally leading to underutilization when reporting rates improve, and overfishing when reporting rates degrade. Thus, while quantifying total catch is necessary for understanding the impact of fisheries on businesses, communities and ecosystems, detecting trends in reporting rates is more important for estimating fishery status and setting sustainable catches into the future.