Impact of WWII and Stormwater Discharges on the Mercury
Status of Fish from Saipan Lagoon, Saipan, CNMI
G R W Denton
1,*
, J O Cruz
1
, M S Trianni
2
, M C Tenorio
3
and R S Masga
4
1
Water and Environmental Research Institute of the Western Pacific, University of
Guam, UOG Station, Mangilao, Guam 96923, USA.
2
National Marine Fisheries Service, Pacific Islands Fisheries Science Center, Saipan
Field Office, Garapan, Saipan, 96950, CNMI
3
CNMI Division of Fish and Wildlife, Caller Box 10007, Saipan, MP 96950
4
CNMI Division of Environmental Quality, P.O. Box 151304, Saipan, MP 96950
Abstract. Total mercury levels were examined in emperor fish (Lethrinus harak and L.
atkinsoni) from 14 coastal sites along the entire length of Saipan Lagoon. These carnivorous
species are abundant throughout the lagoon and are prized tablefish among local residents.
Monitoring their mercu ry content is therefore important fro m a public health standpoint.
Additionally, both species have relatively restricted foraging ranges; hence they are ideal
candidates for delineating areas of mercury enrich ment in their immed iate environ ment. As
expected, relatively high mercury levels were found in specimens fro m the most
industrialized stretch of coastline in the central region of the lagoon. In one notable instance,
however, considerably higher concentrations were encountered in specimens taken from mo re
remotely located waters further south. Surface sediments taken along seaward transects from
the coast in this section, supported the fish data and revealed atypical mercury concentration
profiles that frequently increased with increased distance offshore. These unusual findings
were attributed to the extensive use of mercury in WWII munitions , and to the heavy
bombardment of Saipan during the US invasion of the island in 1944. All evidence in support
of this contention is presented here.
1. Introduction
Saipan is the largest and most densely populated island of the Commonwealth of the Northern
Mariana Island (CNMI). Its economy is primarily tourist-based with visitors to the island largely
attracted by the equable climate, sandy beaches, and extensive coral reefs. Of particular importance
in this regard is Saipan Lagoon, which extends along the western coastline and occupies a total area
of approximately 31 square kilometers. The lagoon harbors a rich diversity of marine life and
supports a variety of commercial and recreational activities[1,2].
Pollution monitoring and assessment studies in Saipan Lagoon have largely been confined to the
central region, which borders the most industrialized part of the island. Heavy metals were among
several recalcitrant chemical groups initially examined in these waters and were subsequently found
to be the contaminants of greatest concern in sediments and biota[3-9]. Waters further south in the
278
Denton, G., Cruz, J., Trianni, M., Tenorio, M. and Masga, R.
Impact of WWII and Stormwater Discharges on the Mercury Status of Fish from Saipan Lagoon, Saipan, CNMI.
In Proceedings of the International Workshop on Environmental Management, Science and Engineering (IWEMSE 2018), pages 278-284
ISBN: 978-989-758-344-5
Copyright © 2018 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved

lagoon, although less obviously impacted by anthropogenic activities, are heavily inundated by
stormwater runoff from commercial and residential premises, highways, and unpaved roads[10].
In 2002, a preliminary analysis of stormwater discharged at several points along the southern half
of Saipan Lagoon revealed surprisingly high levels of mercury[11]. Unlike zinc, copper, and lead,
which correlate strongly with traffic densities and are by far the most prevalent priority pollutants
found in urban runoff[12], mercury has no known source link to vehicle use and is rarely encountered
in stormwater discharges[13]. In USEPA’s 5-year National Urban Runoff Program, for example,
mercury was found in only 9% of the several hundred samples analyzed[12]. This contrasts sharply
with the Saipan study in which mercury was detected 70% of the time. Reported values ranged from
8-150 ng/L and again are remarkable given that detectable levels in runoff seldom exceed 0.01
ng/L[14].
This important study raised the following questions: a) From where was the mercury coming; and
b) what impact was it having on local fisheries? Denton and coworkers, who noted an indiscriminate
scattering of mercury enrichment in sediments throughout the southern half of Saipan Lagoon,
recently proposed a likely answer to the first question[15]. They believed this unusual distribution to
be reflective of the widespread use of mercury in WWII munitions, i.e., as mercury switches in
projectiles and rockets, and mercury fulminate in primers and detonators of artillery shells and
percussion caps of bullets[16,17]. This suggestion is certainly plausible given that the southern
beaches of Saipan Lagoon provided tactical access to the island for US troops during WWII , and that
Japanese defenses positioned along the shoreline were heavily shelled during this maneuver[18,19].
To what extent the aftermath of this historic battle has impacted local fisheries in these waters has
only partially been addressed. In 2004, a fish monitoring program was mounted in the central region
of Saipan Lagoon[7]. Primary anthropogenic disturbances in this region include a commercial port
(Saipan Harbor), a municipal dump (now closed), a power station, a sewer outfall, and two small-
boat marinas. In all, 340 fish representing 67 species from four trophic levels were examined. Total
mercury concentrations in two species of emperor fish (Lethrinus atkinsoni and L. harak) were of
particular interest and are summarized here in Figure 1. The relatively high mercury values noted in
specimens from high disturbance areas (Zones 3-5) were expected, whereas the elevated levels noted
in fish from the most remote site (Zone 7) were not.
A substantial mercury hotspot was subsequently identified in sediments at the northern end of
Zone 7 and is believed to be the source of contamination that impacted these fish[15]. The hotspot
was located in nearshore waters adjacent to the Gualo Rai Road intersection (Figure 1). It extended
seaward from the coast for several hundred meters with levels in excess of 100 ng/g in surface
sediments 100 and 250 meters offshore. Lower, but nonetheless significant levels of enrichment,
were also noted at the 500 meter mark.
Concerns over the possibility of other mercury hotspots impacting fisheries further south in the
lagoon prompted the following investigation to: a) determine mercury levels in soils and sediments
from inland drainage pathways channeling runoff into Zone 7 coastal waters; b) perform additional
mercury analysis on fish from Zone 7, thereby narrowing the focus area for source detection and
delineation; and c) extend the fish monitoring program to the southern end of Saipan Lagoon.
2. Methodology
Surface soil samples were taken for mercury analysis from drainage pathways on the landward side
of the main coastal highway (Beach Road) between Gualo Rai Road and Quartermaster Road. The
primary objective of this part of the investigation was to identify any landward sources of mercury
that could account for the hotspot noted above. Eight stormdrains discharge directly into the ocean
along this 3-km stretch of coastline and include many of the stormwater outlets monitored in
2002[11]. Runoff from this area is primarily derived from Beach Road and adjacent properties within
0.5 km of the coast. It is channeled into the stormdrains via a series of swales and culverts. Surface
Impact of WWII and Stormwater Discharges on the Mercury Status of Fish from Saipan Lagoon, Saipan, CNMI
279

deposits were taken from 29 sampling points along drainage pathways servicing all eight of these
drains. The deposits were scooped up in hand-held, pre-cleaned polypropylene vials and deep frozen
within three hours of collection. The samples ranged in color and texture from light brown, gravelly
sand to darker clay substrates containing varying amounts of organic matter. In the laboratory the
thawed samples were dried to constant weight at 40
o
C and disaggregated by gently kneading between
finger and thumb in clean Ziploc bags. Only the fraction that passed through a 1-mm Teflon screen
was taken for analysis.
Figure 1. A = Map of Saipan (see insert) showing emperor fish capture zones 1-7; roads adjoining
the coastal highway (Beach Road); and the names of coastal villages (italicized). B = Graph of
geometric mean mercury concentrations in axial muscle of fish from each zone (normalized to a
standard 20 cm fork length); whiskers = 95% confidence limits about the mean; numbers in
parentheses = sample sizes.
Fish sampling again focused on L. atkinsoni and L. harak, thereby permitting comparative
analysis with the earlier fish data. These two species are among the most common emperor fish in
Saipan Lagoon[20,21]. They also have restricted foraging ranges, which makes them ideal for
monitoring spatial differences in mercury abundance over relatively short distances. Additionally,
they have similar food preferences, growth rates, and mercury affinities, which supports their use in
tandem for pollution monitoring and assessment studies such as this one. Both species were therefore
opportunistically taken from an additional seven sampling zones between Quartermaster Road and
Agingan Point at the southern end of Saipan Lagoon (Figure 2). Being nocturnal feeders, the fish
were caught at night by hook and line while they were foraging among nearshore seagrass beds.
Captured individuals were placed on ice as soon as possible and transported to the laboratory in
insulated containers. Mercury levels were determined in axial muscle taken immediately below the
dorsal fin of each fish.
All screened surface deposits and fish axial muscle samples were wet digested in a 2:1 nitric-
sulfuric acid mixture at 100
o
C for 3 hours. Analysis was accomplished by cold vapor Atomic
Absorption Spectroscopy using the 'syringe technique' described by Stainton[22]. Calibration
standards (5-20 ng/l) were made up in 10% nitric acid containing 0.05% potassium dichromate as a
(3)
1
2
Saipan Lagoon
Saipan
Susupe
San Jose
Gualo Rai
Garapan
0.000
0.050
0.100
0.150
0.200
0.250
1
2
3
4
5
6
7
Total Hg (µg/g wet weight)
Zone
(21)
(5)
(10)
(22)
(29)
(10)
(3)
A B
Gualo Rai Rd.
Quartermaster Rd.
Microl Rd.
1
3
4
5
6
7
2
IWEMSE 2018 - International Workshop on Environmental Management, Science and Engineering
280

preservative[23]. Approximately 10% of all samples were run in duplicate and were accompanied by
appropriate method blanks and matrix spikes. Accuracy and precision estimates were based on
mercury recoveries from certified standard reference materials and were within acceptable limits.
3. Results and discussion
Mercury levels in all drainage pathway deposits are summarized in Table 1. Overall, levels were
generally low and ranged from 2.45-76.8 ng/g dry weight (average: 17.6 ng/g). Baseline mercury
levels in Saipan soils normally hover around 30 ng/g and rarely exceed 50 ng/g[24]. Clean, sandy
deposits from adjacent nearshore waters are typically an order of magnitude lower[15]. The only
sample to exceed 50 ng/g was retrieved ~100 m inside Pumpkin Street, which runs parallel to a car
dealership and auto repair/storage facilities that contribute runoff to stormdrain 6 (Table 1). The
evidence obtained suggests no major land-based sources of mercury are currently impacting the
lagoon along this stretch of coastline. The results also confirm earlier findings of little more than
light mercury enrichment in beach sediments down gradient of stormdrains discharging into Zone 7
within the road boundaries established above[15].
Table 1. Mercury levels (ng/g dry wt.) in drainage pathway deposits.
Mercury concentrations in the fish examined are summarized in Figure 2 following log-
transformation of the raw data and normalization to a standardized 20-cm fish length. The results are
presented together with the earlier data shown in Figure 1. A comparison between the two datasets
clearly shows that fish from Zones 8-14 consistently contained lower mean mercury concentrations
in their axial muscle when weighed against their northern counterparts. Mercury levels in axial
muscle of fish from non-polluted waters typically range between 0.001-0.100 µg/g wet weight,
depending upon age and trophic level[25]. All recent fish samples analyzed yielded values well
below this upper limit with the exception of one large specimen of L. harak (28.0 cm), which
contained 0.283 µg/g wet weight of mercury in its axial muscle. This particular fish was caught in
waters directly opposite Quartermaster Road (Figure 2). Baseline mercury levels in emperor fish
from reference sites in the northern waters of Saipan Lagoon are around 0.050 µg/g wet weight for a
standardized 20-cm fish[7]. The results from the current study suggest the equivalent benchmark for
fish from the southern half of Saipan Lagoon is about 20% lower. This is indeed good news for
Saipan residents, many of whom regularly fish these waters for subsistence purposes. Nevertheless, a
cautionary note is warranted for consumers of fish above the 20-cm standardized limit upon which
these estimates are predicated. According to USEPA's fish consumption guidelines for the general
population, fish with methylmercury concentrations in their muscle tissue of below 0.088 µg/g wet
(Gualo Rai Road)
1 4 4.49 (2.36-8.67)
2 3 4.03 (3.76-4.29)
3 3 17.7 (7.98-27.3)
4 4 18.6 (9.20-30.1)
5 3 19.0 (13.3-24.6)
6 4 29.2 (5.01-76.0)
7 3 16.9 (6.92-27.0)
8
5 24.6 (11.9-40.5)
(Quartermaster Road)
Stormdrain
Drainage Pathways
Examined
Total Hg (ng/g dry wt.)
Mean (Range)
Impact of WWII and Stormwater Discharges on the Mercury Status of Fish from Saipan Lagoon, Saipan, CNMI
281

weight may be eaten on an unrestricted basis[26]. In contrast, 8-oz fish meals containing the same
concentration as that noted above for the L. harak representative examined should not be consumed
more than twice a week; and not more than three times a month for women of childbearing age,
nursing mothers or sensitive individuals. Mercury in fish occurs predominantly in the highly toxic
methylated form and typically accounts for 80-90% of total mercury in axial muscle tissue[27].
Figure 2. A = Map of Saipan (see insert) showing emperor fishing zones 1-14. Current zone
delineation arrows and numerical markers (8-14) are drawn with white backgrounds. B = current data
plots are shown as white filled circles. All other pertinent map and chart details remain as described
in Figure 1.
4. Conclusions
The collective fish and soil/sediment data gathered during this investigation essentially rules out the
possibility of any significant land-based mercury sources entering the lagoon between Gualo Rai
Road and Microl Road. We are, therefore, of the opinion that the mercury source impacting emperor
fish previously examined from Zone 7, emanates from within the lagoon itself; and is, in all
probability, the mercury hotspot identified earlier and presumed to be a residual munitions artifact
dating back to WWII[15].
Acknowledgements
We are indebted to Dave Benavente (CNMI Division of Coastal Resources Management) for his
assistance with sample collection. This study was made possible through the USGS 104-B Water
Resources Research Institute Program.
0.000
0.050
0.100
0.150
0.200
0.250
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Total Hg (µg/g wet wt.)
Zone
Study Area
1
2
1
Study Area
2
3
Saipan
5
4
Saipan Lagoon
1
2
6
7
9
10
11
12
13
14
Quartermaster Rd.
Gualo Rai Rd.
Microl Rd.
8
Study Area
Agingan Point
(21)
(3)
(5)
(22)
(29)
(10)
(10)
(26)
(10)
(15)
(5)
(7)
(10)
(10)
Lethrinus atkinsoni
A B
San Jose
Gualo Rai
Garapan
San Antonio
Susupe
IWEMSE 2018 - International Workshop on Environmental Management, Science and Engineering
282

References
[1] Doty J E. and Marsh J A Jr 1977 University of Guam Marine Laboratory Technical Report No
33 147 pp
[2] Amesbury S S, Lassuy D R, Myers R F and Tyndzik V 1979 University of Guam Marine
Laboratory Technical Report No. 52 58 pp
[3] Denton G R W, Bearden B G, Concepcion L P, Siegrist H G, Vann D T and Wood H R 2001
University of Guam Water and Environmental Research Institute of the Western Pacific
Technical Report No. 93 110 pp
[4] Denton G R W, Bearden B G, Concepcion L P and et al. 2006 Mar. Pollut. Bull. 52 696
[5] Denton, G R W, Bearden B G and Wood H R 2008 University of Guam Water and
Environmental Research Institute of the Western Pacific Technical Report No. 123 55 pp.
[6] Denton, G R W and Morrison R J 2009 Mar. Pollut. Bull. 58 424
[7] Denton G R W, Trianni M S and Tenorio M C 2010 University of Guam WERI Technical
Report No. 130 98 pp
[8] Denton G R W, Trianni M S, Bearden B G, Houk P C and Starmer J A 2011a Proceedings
2011 International Symposium on Environmental Science and Technology (Dongguan
Guangdong Province China June 1-4, 2011), (Environmental Sciences and Technology, vol.
III) ed S Li, W Wang, P Niu and Y Ann (Science Press USA Inc.) p 983
[9] Denton, G R W, Trianni M S, Bearden B G, Houk, P C and Starmer J A 2011b. J. Toxicol. Env.
Heal. A 74 823
[10] Bearden B G, Houk P C, Bearden C, Chambers D and Simian M 2010 Final Commonwealth of
the Northern Mariana Islands Integrated 305(b) and 303(d) Water Quality Assessment
Report. (Division of Environmental Quality) revised November 2010 106 pp
[11] Environet Inc. 2007 Draft Environmental Restoration Report, Aquatic Ecosystem Restoration
Study, at Saipan, Commonwealth of Northern Marianas Islands. Prepared for US Army
Core of Engineers, Honolulu Engineering District by Environet Inc. Honolulu, Hawaii
(contract no. DACA83-00-D-0037) February 2007 126 pp
[12] USEPA 1983 Results of the Nationwide Urban Runoff Program (volume 1) U.S.
Environmental Protection Agency, Water Planning Division, NTIS No PB84-185552,
Washington DC
[13] Fulkerson M, F N Nnadi and L S Chasar 2007 Water Soil Air Poll. 185 21
[14] Morace J L 2012 U.S. Geological Survey Scientific Investigation Report 2012-5068, 68 pp
[15] Denton, G R W, Emborski C A, Habana N C and Starmer J A 2014. Mar. Pollut. Bull. 81 276
[16] US Navy 1946 VT Fuses for Projectiles and Spin-stabilized Rockets. Navy Department,
Bureau of Ordnance Washington 25, D.C. Publication No. OP 1480 (first revision) 15 May
1946 42 pp
[17] US Navy (1947). U.S. Explosive Ordnance. Navy Department, Bureau of Ordnance
Washington, D.C. Publication No. OP 1664 (vol. 1), 28 May 1947, 581 pp
[18] Crowl P A 1960 War in the Pacific: Campaign in the Marianas. U.S. Army Center of Military
History Publication 5-7, World War II 50
th
Anniversary Commemorative Edition 1993 (U.S.
Government Printing Office), 525 pp. Available at:
https://history.army.mil/html/books/005/5-7-1/CMH_Pub_5-7-1.pdf
[19] Trueman C 2000 The Battle of Saipan 1944, History Learning Site. Available at
http://www.historylearningsite.co.uk/battle_saipan_1944.htm
[20] Graham T 1994 Biological Analysis of the Nearshore Reef Fish Fishery of Saipan and Tinian.
CNMI Division of Fish and Wildlife Technical Report 94-02
[21] Taylor B M and J L McIlwain 2010 Mar. Ecol-Prog. Ser. 411 243
[22] Stainton M P 1971 Anal. Chem. 43 625
[23] Feldman, C 1974 Anal. Chem. 46 99
Impact of WWII and Stormwater Discharges on the Mercury Status of Fish from Saipan Lagoon, Saipan, CNMI
283

[24] Denton G R W, Emborski C A, Hachero A A B, Masga R S and Starmer JA 2016 Environ. Sci.
Pollut. R. 23 11339
[25] Holden A 1973 J. Food Technol. 8 1
[26] USEPA 2000 Guidance for Assessing Chemical Contaminant Data for Use in Fish Advisories:
Volume 2, Risk Assessment and Fish Consumption Limits (3
rd
edition), United States
Environmental Protection Agency, Office of Water, Washington DC, Document No. EPA
823-B-00-008
[27] Storelli M M, Storelli A, Giacominelli-Stuffler R and Marcotrigiano G O 2005 Food Chem. 89
295
IWEMSE 2018 - International Workshop on Environmental Management, Science and Engineering
284