Five Days in June:  

The Archaeological and Historical  

Legacy of Saipan’s Open Ocean Seadrome 

By  

Geoffrey M. Anthony 

May, 2025 

Director of Thesis: Dr. Jennifer F. McKinnon, Ph.D. 

Major Department: Department of History 

ABSTRACT 

 Saipan’s open ocean seadrome was an unconventional and expedient solution to an 

emergent operational challenge. Operation FORAGER, the invasion of the Northern Marianas, 

was unexpectedly contested by the Japanese First Mobile Fleet. The United States Fifth Fleet 

immediately implemented an expedient non-traditional seadrome in the open ocean for emergent 

long-range searches. It accepted and managed significant operational risk by establishing and 

operating this atypical seadrome outside a sheltered anchorage — and in harm’s way — to 

support vital PBM-3D Mariner seaplane night Radio Direction and Ranging search operations. 

Previous archaeological efforts have yielded a commanding knowledge of the traditional 

seadrome in Saipan’s Tanapag Harbor. Surveys of the captured Japanese seaplane base with 

existing infrastructure have contributed markedly to the archaeological record. A more thorough 

effort devoted explicitly to the innovative, high-risk, high-reward implementation of the initial 

open ocean seadrome and its contribution to the archaeological record has yet to be undertaken.  

This open ocean seadrome, established in the opening days of the battle, provided the 

initial long-range search crucially required for the security of the United States Fifth Fleet. 

Assessing the impact these patrol planes had on the conduct and outcome of the battle informs an 



   

 

 
 

underreported, yet critical, aspect of the Battle of the Philippine Sea. This facility also served as a 

proof of concept that informed future Pacific seadrome requirements.  

Although the exact location of this seadrome remains speculative, archival resources 

provide the level of geographic fidelity needed to conduct an appropriate archaeological site 

survey. A PBM-3D maritime patrol seaplane accidentally sank at this anchorage, but without the 

catastrophic damage and dislocation generally associated with traditional aircraft crashes. 

Locating, surveying, and recording this aircraft in situ would theoretically provide consequential 

site formation and cultural and environmental impact data.  

Other archaeologically significant artifacts may also be present. These artifacts would 

further contribute to and validate the archaeological record and inform open ocean seadrome 

operational practices and seaplane tender support activities. The open ocean seadrome was 

established upon the Saipan Shelf, the invasion's primary amphibious force staging area. The 

potential for opportunistic incidental artifact discovery should not be underestimated.  

This area is approximately 100 feet (31 meters) deep. Search efforts included traditional 

methods and sought to inform the viability and reliability of leveraging various detection 

technologies, such as Light Detection and Ranging, adapted for maritime archaeological surveys. 

Geographic Information Systems software and data were utilized to establish search areas and 

parameters based on historical records and to assist in mapping the locations of any cultural 

material associated with the open ocean seadrome. Experimentation with these technologies 

contributes to their continued development and integration into existing archaeological methods, 

improving their future effectiveness.  

Key Words: Saipan, Battle of the Philippine Sea, seadrome, seaplane anchor, seaplane tender, 

bowser boat, patrol aircraft, Martin PBM Mariner, Operation FORAGER, VP-16, Pacific Theater, Second 

World War.   



   

 

 
 

  



   

 

 
 

Five Days in June:  

The Archaeological and Historical  

Legacy of Saipan’s Open Ocean Seadrome 

 

 

 

A Thesis 

Presented to the Faculty of the Department of History 

East Carolina University 

 

 

In Partial Fulfillment of the Requirements of the Degree 

Masters of the Arts in Maritime Studies 

 

 

By 

Geoffrey Matthew Anthony 

Colonel, United States Marine Corps, Retired 

 

May 2025 

 

 

 

Director of Thesis: Dr. Jennifer F. McKinnon, Ph.D. 

Thesis Committee Members: 

Dr. M. Todd Bennett, Ph.D. 

Dr. Jason T. Raupp, Ph.D.   

 

 

 

 

 



   

 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

© Geoffrey M. Anthony, 2025

 

 

 

 

 

 

 

 

 

 

 

 

 

 



   

 

 

DEDICATION 

Operation FORAGER was the largest operation of the Pacific War at the time it was 

executed. The tyranny of geography squarely favored the Japanese with their interior lines of 

communications and forward bases close to the Marianas. Japanese facilities in the Marianas and 

at Truk, Yap, and Palau could support Japanese efforts to contest U.S. operations. Consequently, 

these bases had to be neutralized.  

The FORAGER operating area fell squarely in Pacific Fleet Commander Admiral Chester 

Nimitz’s Central Pacific Area of responsibility. Yet, in a display of interservice cooperation, 

much of the responsibility for neutralizing these bases fell to General Douglass MacArthur’s 

Army Air Forces operating from Southwest Pacific Area bases. The 13th Air Force had recently 

occupied an advanced base in Los Negros, Admiralty Islands, just north of New Guinea. Four 

squadrons of heavy bombers from the 307th Bomb Group (Heavy), the fabled Long Rangers, a 

moniker earned after their extended long-range raids over the open ocean against Wake Island 

early in the war, began combat operations from there in May 1944. 

The 307th participated in attacks on Truk, Yap, Palau, and other Japanese bases to 

neutralize them in the days and weeks leading up to Operation FORAGER. Their B-24 

Liberators were later called upon to fly 1,200 miles (1,931 kilometers) search routes to help 

locate the recently sortied Japanese First Mobile Fleet to protect the invasion force’s southern 

flank once the Battle of Saipan commenced on 15 June 1944. Following the defeat of the 

Japanese Navy during the Battle of the Philippine Sea, the Long Rangers returned to attacking 

targets in the Caroline Islands in continued support of Operation FORAGER and in anticipation 

of future offensive operations.  



   

 

 

On 10 August 1944, the same day the Battle for Guam concluded, ending the Southern 

Mariana Islands campaign, a B-24J (44-40571) from the 371st Bomb Squadron, 307th Bomb 

Group (Heavy), 13th Air Force took flight from Mokerang Airdrome, Los Negros, Admiralty 

Islands to deliver a strike against Colonia, the capitol of Yap Island. Separated from its flight by 

weather en route, it still pressed the attack after a belated arrival at the target. It did not return. 

Post-war research and interviews suggest it was destroyed over the target area by anti-aircraft 

fire. The Japanese plucked two surviving crew members from the island’s barrier reef, although 

they went missing during their captive transit and remain unaccounted for. The other nine 

members of the crew, including an observer who was also flying that day, are presumed to have 

gone down with the Liberator, reported by witnesses to have crashed into Colonia Harbor, where 

they remain.  

Although this scene played out in various forms countless times throughout the war, this 

specific event holds special meaning to the author. The pilot of this Liberator was 24-year-old 

Don Alpheus Anthony from Punxsutawney, Pennsylvania. This thesis is therefore dedicated in 

memory of my grandfather, First Lieutenant Anthony, and his crew, killed in action on 10 August 

1944 while conducting combat operations in the Caroline Islands (FIGURES i, ii, & iii). 

Although you remain far from home, you are not forgotten.  

  



   

 

  

FIGURE i (top): Second Lieutenant Don A. Anthony, United States Army Air Force, (top left) and crew 

posing in front of the 307th Bomb Group’s storied B-24D Frenesi (42-40323), the first Pacific Theater 

bomber to reach 100 combat missions. Lieutenant Anthony piloted this bomber on its 98th mission, 

sustaining battle damage that required landing with a destroyed main mount landing gear. (bottom): 

Lieutenant Anthony (bottom left), and crew in front of a B-24 bomber (Photos courtesy Anthony family). 



   

 

 

 

FIGURE iii: Aviation Cadet Don. A. Anthony, U.S. Army Air Force. (Photo courtesy Anthony family). 



   

 

 

ACKNOWLEDGEMENTS 

 I want to thank the East Carolina University Department of History and Department of 

Maritime Studies staff. I welcomed with enthusiasm the level of dedication and professionalism 

exhibited by the whole of the academic cadre. I must single out for gratitude, first and foremost, 

Dr. Jennifer McKinnon. My first unknowing exposure to her work was when I purchased a copy 

of Underwater Archaeology of a Pacific Battlefield: The WWII Battle of Saipan. As an interested 

career Marine and aspiring maritime archaeologist, I remember coming away overwhelmingly 

impressed, yet wondering who would put this level of Yeoman’s work into a relatively obscure 

battle so central to United States Marine Corps lore. I count myself lucky to have found out 

firsthand. 

 As my thesis advisor, Dr. McKinnon has shown unrelenting patience, allowing me 

extraordinary latitude as a non-traditional student. She permitted me to explore flights of fancy in 

my research and thesis writing before serving as the needle to the Pole, and gently guided me 

back to True North. As a mentor and Defense Prisoner of War/Missing in Action Accounting 

Agency project lead, she exhibited all the leadership traits and management skills that would 

have made any Marine Corps officer proud. She has my eternal thanks.   

 Dr. Jason Raupp, a thesis committee member with whom I had too few classes, also 

provided exceptional guidance in my studies. He was a steady hand, a continuous source of 

enthusiasm for the community, and a tremendous collaborator as I progressed through the 

program. Dr. M. Todd Bennett, who graciously offered to serve on my thesis committee despite 

limited contact with me, is a source of inspiration for his professional accomplishments. His love 

and command of history and prolific publishing record serve as an inspiration for my future. 

Finally, Dr Nathan Richards, the Chair of the Department of Maritime Studies and instructor 



   

 

 

extraordinaire, typifies what it is to be an underwater archaeologist. His sage counsel and eternal 

availability to his students for all things, big and small, have made him a community pillar. I am 

grateful to have had the opportunity to receive his tutelage.  

 I would be remiss if I did not thank others who have made the program the envy of the 

maritime studies world. Mark Keusenkothen and Ryan Bradley introduced me to a more 

disciplined, scientific, and safe diving methodology, doing their best and mostly succeeding in 

breaking my many decades-old habits. The program would not be what it is if not for their 

efforts.  

I am obliged to Aleck Tan. Without her assistance and expertise in Light Detection and 

Ranging, and Geographic Information Systems, survey efforts in Saipan would not have 

occurred effectively.  

Finally, I would like to express my heartfelt appreciation to my fellow students for 

tolerating a “peer” thirty years their senior. Their willingness to assist me with my many 

technical challenges made my progress in the program much more manageable. I am heartened 

to see their competence and commitment, and I am assured that the community's future is in 

good hands. 

 

 

 

 

 

  



   

 

 

TABLE OF CONTENTS 

TITLE………………………………………………………………………………………...……i 

COPYRIGHT………………………………………………………………………………….….ii 

DEDICATION…………………………………………………………………………………....iii 

ACKNOWLEDGEMENTS…………....………………………………………………………...vii 

TABLE OF CONTENTS…………………………………………………………………..……..ix 

LIST OF FIGURES……………………………………………….…………………......……....xii 

LIST OF ABBREVIATIONS…………………………………………………………………....xv 

PROLOGUE……………………………………………..……………..……………..….…...xviii 

CHAPTER 1: INTRODUCTION…………………………………………………………………1 

1.1. HISTORY OF SAIPAN……………………………………………………………….1 

1.2. PREVIOUS ARCHAEOLOGICAL EFFORT………………………………………..7 

1.3. LITERATURE REVIEW……………………………………………………………..7 

1.4. RESEARCH AIMS AND QUESTIONS…………………………………………….13 

1.5. ARCHIVAL RESEARCH…………………………………………………………...15 

1.6. TERMINOLOGY……………………………………………………………………17 

CHAPTER 2: HISTORY………………………………………………………………………...20 

2.1. PBM-3D MARINER………………………………………………………………..20 

2.2. PATROL SQUADRON 16 (VP-16)……………………………………………...….29  

2.3. SEAPLANE TENDERS…………………………………………………………….31 

2.3.a.USS Ballard (AVD-10)…………………………………………………….32 

2.3.b. USS Onslow (AVP-48)……………………………………………………35 

2.3.c. USS Pocomoke (AV-9)…………………………………………………….37 



   

 

 

2.4. SAIPAN’S OPEN OCEAN SEADROME…………………………………………..40  

2.4.a. Planning…………..……………………………………………….……….41 

2.4.b. Establishment and Operations of the Open Ocean Seadrome...…...…..….42 

2.4.c. Loss of BuNo 48173…………...………………………………………….63 

2.4.d. Endgame………………...…………………………………………...……65 

CHAPTER 3: THEORY………………………………………………………………..………..69 

3.1. THEORETICAL CONSIDERATIONS…………………….……………………….69 

3.2. PBM-3D BUNO 48173: A PREDICTIVE SITE FORMATION ASSESSMENT…73 

3.3. PBM-5A BUNO 59127: A CAUTIONARY TALE………………………..………77 

3.4. IN SITU CONSERVATION: THE PATH FORWARD……………………………79 

CHAPTER 4: RESEARCH METHODOLOGY………………………….……………………..81 

4.1.HISTORICAL AND ARCHIVAL RESEARCH………………..……….………….82 

4.2.BATTLEFIELD EVALUATION AND ASSESSMENT……………..…………….86 

CHAPTER 5: ARCHAEOLOGICAL ACTIVITIES………………………………………..…100 

5.1. LIDAR AND ERSI ARCGIS EXPERIMENTATION…………………………..…101 

5.2. TOW BOARD SURVEY………………………………………………..…………105 

5.3. IDENTIFICATION METHODOLOGY………………………………..……….…114 

5.4. FUTURE ARCHAEOLOGICAL ACTIVITIES…...………………………………116 

CHAPTER 6: DISCUSSION………………………….………………………………………..122 

6.1 PRIMARY RESEARCH QUESTION………………..……..……………………..122 

6.2.SECONDARY RESEARCH QUESTIONS……………………………….……….131 

CHAPTER 7: CONCLUSIONS………………………………………………………………..144 

WORKS CITED…………………………………………………………………………...…...146 



   

 

 

APPENDICES 

A. ANATOMY OF A FRIENDLY FIRE ENGAGEMENT……………………………....159 

B. ADDITIONAL ARCHAEOLOGICAL POTENTIAL…………………………………164 

  



   

 

 

LIST OF FIGURES 

 

FIGURE i: Second Lieutenant Don A. Anthony and crew in front of B-24DFrenesi ………..….vi 
 

FIGURE ii: Second Lieutenant Don A. Anthony and crew……………………………….….......vi  

 

FIGURE iii: Aviation Cadet Don A. Anthony…………………………………….………….….vii 

 

FIGURE 2.1: Wright 2600-14 engine…………………………………………….……...............22 

 

FIGURE 2.2: Schematic of the PBM-3D ………….……………………………………………25 

 

FIGURE 2.3: Three view drawings for PBM-3D ………….……………………………...…….27 

 

FIGURE 2.4: 1942 World War II Martin Aircraft PBM-3 propaganda poster...............................28  

 

FIGURE 2.5: Lieutenant Commander William J. Scarpino on 13 February 1945……….……...30 

 

FIGURE 2.6: VP-16 PBM-3D Mariner.........................................................................................31 

 

FIGURE 2.7: USS Ballard (DD-267) in 1920...............................................................................33 

 

FIGURE 2.8: Northill print advertisement for its compact, folding seaplane anchor……….......34 

 

FIGURE 2.9: USS Ballard (AVD-10) at Mare Island Navy Yard 30 December 1943..................35  

 

FIGURE 2.10: USS Onslow (AVP-48) launch at the Lake Washington Shipyards……………...36 

 

FIGURE 2.11: Side view of USS Onslow (AVP-48) at Saipan Island in the Marianas................37 

 

FIGURE 2.12: SS Exchequer in Pascagoula, Mississippi 1940....................................................38 

 

FIGURE 2.13: Diagram of large seaplane tender deck configurations.........................................39 

 

FIGURE 2.14: USS Pokomoke (AV-9) at Kwajalein, Marshall Islands........................................40 

 

FIGURE 2.15: The bombing of the Japanese seaplane base at Flores Point, Saipan…….……...42 

 

FIGURE 2.16: Diagram showing air search fans for land-based air searches on 18 June 1944...55 

 

FIGURE 2.17: Japanese Kawanishi H8K “Emily” flying boat.....................................................57 

 

FIGURE 2.18: Damage to aircraft 16-P-1 documented while aboard Pokomoke.........................58 

 

FIGURE 2.19: PMB-3D 16-P-1 under tow to Pokomoke..............................................................59 

 

FIGURE 2.20: PMB-3D 16-P-1 being craned aboard Pokomoke.................................................59 



   

 

 

FIGURE 2.21: Motor launch with deceased Aviation Machinist Mate Second Class Person......60  

 

FIGURE 2.22: Aviation Machinist Mate Second Class Person’s remains aboard Pokomoke......60 

 

FIGURE 2.23: 19 June 1944 chart plot from CTG 58.1’s Combat Information Center...............61 

 

FIGURE 2.24: Navy PBM-3 Martin “Mariner” Patrol Bomber recognition poster......................62 

 

FIGURE 2.25: USS Pokomoke lowers #1 aviation fuel barge into the water................................64 

 

FIGURE 2.26: Vice Admiral John Hoover, Commander, Forward Area......................................66 

 

FIGURE 2.27: Saipan seadrome as seen from USS Pokomoke……………………………..…..68 

 

FIGURE 3.1: Aircraft site formation processual model…………………………………………73 

 

FIGURE 4.1: USS Maryland (BB-46) torpedo attack diagram…………………………..……...84 

 

FIGURE 4.2: Anchoring plan from the Operation FORAGER Training Plan..............................85 

 

FIGURE 4.3: Japanese 25mm anti-aircraft gun ………………………………………..………..91 

 

FIGURE 4.4: Japanese Type 10 120mm Dual Purpose gun emplaced in Saipan………..……...92 

 

FIGURE 4.5: Anti-Aircraft Action Report and Diagrams (USS Hercules)..................................95 

 

FIGURE 4.6: Anti-Aircraft Action Report and Diagrams (USS Ballard)....................................96 

 

FIGURE 4.7: Hydrographical Survey Plan for Operation FORAGER.........................................97 

 

FIGURE 4.8: PBMs of VP-16 in formation en route to Palau Islands from Saipan…………….99 

 

FIGURE 5.1: Targeted dives and visual tow board survey area on the Saipan Shelf..................101 

 

FIGURE 5.2: Target locations developed by Aleck Tan using NOAA LiDAR...........................102 

 

FIGURE 5.3: Reported location of PBM-3D BuNo 48173 and Anchorage A-13.......................104 

 

FIGURE 5.4: ArcGIS generated plot track..................................................................................105 

 

FIGURE 5.5: Landing craft ramp………………………………………………………………111 

 

FIGURE 5.6: Marine inline engine……………………………………………………………..111 

 

FIGURE 5.7: Drive shaft associated with marine inline engine………………………………..112 

 



   

 

 

FIGURE 5.8: UXO located in survey area…………………………………………………..…112 

 

FIGURE 5.9: Actual tow board search tracks 9–11 May 2024....................................................114 

 

FIGURE 5.10: SEARCH 2008 survey of Tanapag Lagoon…………..……………….……..…118 

 

FIGURE 5.11: Proposed survey of the Saipan Shelf with overlays…………....………………120 

 

FIGURE 5.12: Proposed survey of the Saipan Shelf without overlays….………..……..……..120 

 

FIGURE 5.13: Proposed survey of the outer reef area with overlays………………………….121 

 

FIGURE 5.14: Proposed survey of the outer reef area with overlays……….……….………..121 

 

FIGURE 6.1: Opening of the Tanapag Harbor seaplane base on 10 July 1944…………..….…125 

 

FIGURE 6.2: USS Pokomoke (AV-9) receiving PBM spare parts...............................................132 

 

FIGURE 6.3: USS Green (AVD-13) diagram on refueling/rearming techniques…...................133 

. 

FIGURE 6.4: Anchoring system used U.S. PBMs North Atlantic..............................................138 
 

FIGURE 6.5: Schematic of a 30-foot U.S. Navy motor launch..................................................141 

 

FIGURE 6.6: Schematic of a 30-foot U.S. Navy motor launch..................................................141 

 

FIGURE A.1: USS Belleau Wood (CVL-24) underway on 22 December 1943.........................157 

 

FIGURE A.2: VF-24 squadron photo taken while embarked aboard USS Belleau Wood..........161 

 

FIGURE B.1: The Saipan landing plan.......................................................................................162 

 

FIGURE B.2: “D-Day at Saipan” watercolor by Staff Sergeant John Fabion.............................163 

 

FIGURE B.3: Salvage of small craft damaged during the initial invasion of Saipan.................164 

 

FIGURE B.4: LVTs disabled at the water’s edge........................................................................166 

 

FIGURE B.5: 5th Amphibious Corps reporting on LVT losses during the Battle of Saipan…..167 

 

FIGURE B.6: Japanese torpedo bomber impacting USS Mercury’s cargo boom……….……..171 

 

FIGURE B.7: UDT conducting explosive demolition operations on Saipan’s reefs….………..173 

  



   

 

 

LIST OF ABBREVIATIONS 

 
AAF  Army Air Force 

AK  Cargo Ship 

AKA  Attack Cargo Ship 

AK  Cargo Ship 

AP   Transport Ship 

APA  Attack Transport Ship 

ASW   Anti-Submarine Warfare 

AUV  Autonomous Underwater Vehicle 

AV  Large Seaplane Tender 

AVD  Destroyer Seaplane Tender 

AVP  Small Seaplane Tender  

BB  Battleship 

CAP  Combat Air Patrol 

CIC  Combat Information Center 

CNMI   Commonwealth of the Northern Marianas  

CTG  Commander, Task Group 

CVL  Aircraft Carrier, Light 

DD  Destroyer 

DE  Destroyer Escort 

DPAA  Defense POW/MIA Accounting  

DUKW 2.5-ton Six-Wheel Amphibious Truck (Based upon Manufacturing Code) 

ECU   East Carolina University  



   

 

 

FDC  Fighter Direction Center 

FPAN   Florida Public Archaeology Network 

GIS  Geographic Information Systems  

HF  High Frequency 

HPO   Historic Preservation Office  

IFF  Identification, Friend or Foe 

IJN  Imperial Japanese Navy  

JATO  Jet Assisted Take Off  

LiDAR  Light Detection and Ranging  

LCC  Landing Craft, Control 

LCDR  Lieutenant Commander 

LCI  Landing Craft, Infantry 

LCVP  Landing Craft, Vehicle, Personnel 

LOD  Line of Departure 

LSD  Landing Ship, Dock 

LST  Landing Ship, Tank 

LT   Lieutenant 

LVT  Landing Vehicle, Tracked 

LVT(A4) Landing Vehicle, Tank 

MDSU  Mobile Diving Salvage Unit  

NAAS  Naval Auxiliary Air Station 

NAS  Naval Air Station 

NATS  National Air Transport Service 



   

 

 

NHHC  Naval History and Heritage Command  

OSU2   Observation Seaplane, United Aircraft Division 

PATSU  Patrol Aircraft Service Unit  

PBEC  Pacific Basin Environmental Consultants 

PPC  Patrol Plane Commander 

RADAR Radio Direction and Ranging 

RADM Rear Admiral  

ROV  Remotely Operated Vehicle 

SOPA  Senior Officer Present Afloat 

SS  Submarine 

TFD  Task Force Dagger 

UDT  Underwater Demolitions Team 

UN  United Nations 

U.S.  United States 

UXO  Unexploded Ordnance 

VADM  Vice Admiral  

VF  Fighter Squadron  

VP  Patrol Squadron 

  



   

 

 

PROLOGUE 

On 4 May 1944, submarine USS Flying Fish (SS-229) efficiently slipped its berth in 

Mujaro, Marshall Islands, escorted out of the harbor by the new destroyer escort USS Manlove 

(DE-36), and proceeded west toward the Philippines. So began the veteran boat’s tenth war 

patrol. Six weeks later, it was on patrol 12 miles (20 kilometers) north northeast of the San 

Bernadino Straits lighthouse. This was an optimal position to detect and intercept vessels using 

this route to transit from the relatively enclosed Samar Sea to the open Philippine Sea to the east. 

Its placement was intentional (Flying Fish 1944:2).  

As the United States prepared to execute Operation FORAGER, the invasion of the 

Marianas Islands, they needed reliable information on any Japanese naval efforts to contest them. 

The Navy had set up a robust submarine picket line to collect this vital information. Flying Fish, 

part of this picket line, detected heightened Japanese naval air patrol activity near the Straits at 

0630 on 15 June 1944. Nine hours later, the submarine observed two Jake (Aichi E13A) long-

range patrol seaplane aircraft “suspiciously staying to the SE near Biri Head on the east 

channel.” Hedging east in their direction, it soon observed additional aircraft, possibly carrier-

based, at 1633. Two masts were spotted through the periscope just thirteen minutes later. Flying 

Fish closed the distance at high speed to establish the identity of the surface contacts.  

The faraway masts quickly developed into a large Japanese task force, assessed as an 

“estimated three carriers, three battleships, several cruisers and destroyers”(Flying Fish 

1944:19). The enemy fleet was heading 080°, almost directly away from the U.S. submarine, at 

20 knots. Flying Fish could not develop an attack solution or even shadow this enemy force 

because of the shallow water (<100 ft./31 m) and even shallower nearby banks. Regardless, the 

submarine surfaced and sent a contact report an hour after losing contact (Flying Fish 1944:19).  



   

 

 

This report, in conjunction with a previous sighting by USS Redfin (SS-272) and later 

reports by USS Cavella (SS-244) and USS Sea Horse (SS-304), began a series of events that 

culminated in one of the most significant naval actions of the war (Morison 1953:237, 241). The 

Battle of the Philippine Sea demolished what was left of the Japanese naval air forces, further 

hobbled the Japanese Combined Fleet, and paved the way for follow-on U.S. operations against 

the Empire of Japan. This action left an extraordinary historical legacy that has been studied and 

evaluated even before the after-action report ink dried. It also left an unparalleled archaeological 

legacy that has only received similar attention in recent decades, much of which still awaits 

discovery.



   

 

 

CHAPTER 1: INTRODUCTION 

Before beginning this archaeological and historical journey, it is essential to establish a 

baseline knowledge regarding the island of Saipan, previous archaeological activities, and 

existing resources available to the contemporary scholar. More important is to present what and 

why questions this thesis seeks to answer – essentially its raison d’être. These issues, and more, 

are addressed presently.  

1.1. HISTORY OF SAIPAN 

Saipan, a once peaceful and nondescript Pacific Island, was thrust into the global 

spotlight by the great contest between Japan and the United States (U.S.) It became one of the 

most brutal killing fields of a particularly vicious war, upon which hinged hope for ultimate 

victory by both belligerents. The Japanese viewed it as part of the “Absolute Defense Sphere” 

that must be defended at all costs (Drea 1998:187). The U.S. recognized it as the key to 

breaching the inner Japanese defense perimeter, with Admiral Ernest King, Commander in Chief, 

U.S. Fleet and Chief of Naval Operations himself, championing the plan to seize them, arguing, 

“The Marianas are the key of the situation because of their location on the Japanese line of 

communication” (Crowl 1960:6–7). Admiral King found an unlikely ally in the fledgling Army 

Air Force (AAF), which desired Saipan as a base for strategic bombing. AAF planners asserted, 

“plans for the acceleration of the defeat of Japan would place emphasis upon the seizure of the 

Marianas at the earliest possible date, with the establishment of heavy bomber bases as the 

primary mission” (Crowl 1960:12).  

Although the Marianas gained newfound significance due to their central strategic 

location in the Pacific, they had been under colonial occupation since they were “discovered” by 

Ferdinand Magellan in 1521. The island group was alternately known throughout history 



   

 

2 
 

originally as “Islands of the Lateen Sails,” then derisively as “Islands of the Thieves” due to 

misunderstandings regarding the communal view the Chamorro peoples held regarding personal 

property. Then, finally, as “The Marianas,” today’s naming convention, in tribute to Spain’s 

Queen Mariana (Hallas 2019:4–5). Spain spent the three hundred years following Magellan’s 

discovery primarily concerned with converting the Chamorro people to Catholicism. This 

process, combined with disease and draconian suppression of rebellion, dramatically reduced the 

original inhabitants' numbers from 45,000 to just 5,000 by 1693 (Denfeld 1997:1).  

After the U.S. defeated Spain in the Spanish-American War, Guam was occupied as a 

coaling station in 1898. Spain was allowed to sell the rest of the Marianas to Germany for $4.5 

million (Hoffman 1950:12). German control was quickly extinguished after the Japanese 

declared war on them in 1914 during World War I as a pretext to seize German possessions in the 

Pacific. Japanese possession was validated in 1920 by a mandate from the United Nations (UN), 

allowing for their administration of all islands in the Marianas group except for Guam, which the 

U.S. still occupied. Japanese occupation revolutionized the sugar industry in Saipan and 

witnessed a massive influx of Japanese citizens, especially farmers from Okinawa. Thus, 

The population in 1936 was reported to be about 23,800, the greater part being Japanese with the remained 

mostly Chamorros, Koreans, and Kanaka natives. In 1941, the civilian population was estimated to be 

between 30,000 and 35,000. Garapan, the principal town located on the west coast just south of Tanapag 

Harbor, was estimated to have a population of about 10,000. (CINCPAC 1944:46) 

 

Beyond this broad pre-war historical overview, the optimal way to describe and assess the 

Marianas, and Saipan in particular, is through the eyes of the U.S. operational planners to 

highlight their importance to both combatants. The Marianas Group, of which Saipan is just one 

of 15 islands, extends 450 miles (mi.) (724 kilometers [km]) from north to south in an evenly 

shaped arc. The northern islands, except for Pagan, were of limited military significance. The 

southern islands are raised coral formations with volcanic cores, consisting of Saipan, Tinian, 



   

 

3 
 

Rota, and Guam, and were “important bases in the Japanese island defenses and 

communications” (CINCPAC 1944:45). Saipan, the second largest island in the Marianas Group, 

was the principal Japanese position in the area and was thus selected as the first objective for 

capture. This island is approximately 12.5 mi. (20 km) long, oriented from the northeast to 

southwest, and has a maximum width of 5.5 mi. (9 km) (CINCPAC 1944:44–45). Although 

wartime planners believed it encompassed approximately 72 mi.2 (116 km2), a more accurate 

contemporary assessment is 47 mi.2 (76 km2) (Hallas 2019:4).   

The terrain is generally mountainous but has considerable arable land dedicated to sugar 

cultivation. The critical terrain of the island is a mountain ridge stretching about 7.5 mi. (12 km) 

down the center of the island. Mount Tapotchua is the most prominent feature, located almost in 

the center of the mountain ridge, its highest point towering 1,554 ft. (474 m) above the island. 

The sides of the northern mountain ridge are broken up into numerous terraces separated by 

rocky cliffs. Abundant ravines characterize the southern mountain ridge. Caves and mountain 

caverns are ubiquitous. Moving further south and southeast away from Mount Tapotchua, 

mountain ridges give way to the flat plateaus, punctuated by several steep breaks, that comprise 

the island’s eastern peninsula. On the west side of the island, between Garapan and Agingan 

Point, rests a low and poorly drained coastal plain with a small lake and swampy ground. Areas 

of land not being cultivated were generally wooded or covered in scrub (CINCPAC 1944:45–46). 

Saipan’s approximate 40 mi. (64 km) of coastline is characterized by steep, rocky cliffs 

on the north, east, and south coasts and is unsuitable for landings. In contrast, the island's 

western side has sandy beaches for most of the island’s length, with gently sloping terrain 

extending 500 to 5,000 yd. (457 to 4,570 m) inland. A protective barrier reef and prevailing 

easterly trade winds ensured generally light surf conditions on the western coast. Beyond the 



   

 

4 
 

barrier reef is a shallow shelf that extends a considerable distance offshore. The Saipan Shelf, 

with depths of less than 20 fathoms (ftm.) (120 ft./37 m), provided an exceptional anchorage and 

operating area for a substantial number of ships (CINCPAC 1944:46–47). Unanticipated by U.S. 

Navy planners, it also served as a mooring area for PBM-3D Mariner seaplanes.  

As the clouds of war gathered in the 1930s, much of the Japanese effort shifted from 

sugar cultivation to military infrastructure, with a heavy emphasis on airfield construction 

development in defiance of their UN mandate (Denfeld 1997:6). Consequently, on the eve of the 

invasion, there were three airfields on Saipan. Aslito Field in the south was the most important 

and capable, with hangars and major repair facilities. Two other strips, one in the west near 

Chalan Kanoa and one in the north near Marpi Point, were small with limited facilities. The 

Japanese military also realized the potential for seaplane operations, and constructed a seadrome 

in Tanapag Harbor on the west coast, complete with two seaplane ramps, barracks, shops, 

warehouses, and three steel hangers (Denfeld 1997:9). This small harbor could accommodate as 

many as twelve ships or more and served “as a fueling and supply station for ships en route to 

and from the Empire” (CINCPAC 1944:46). 

Intelligence estimates as late as 29 May 1944 identified “12 coastal defense guns, 5 dual-

purpose single mounts, 30 or more heavy anti-aircraft guns, 90 or more medium anti-aircraft 

guns, and numerous machine guns” (CINCPAC 1944:47). Post-battle assessment later 

determined this represented just a fraction of the heavy artillery Japan had at its disposal on 

Saipan. Only a third of their heavy weapons were mounted and prepared for combat; the rest 

were victims of the earlier-than-anticipated arrival of the invasion and an effective submarine 

campaign that choked Saipan of desperately needed defensive construction materials (CINCPAC 

1944:46). U.S. planners underestimated the number of Japanese defenders. They were surprised 



   

 

5 
 

to learn later that 29,702 Japanese servicemen from all branches were present to contest their 

conquest of Saipan (Hoffman 1950:17). 

In June 1944, a supporting campaign was aggressively implemented to isolate the 

Marianas and disrupt or destroy enemy land-based air capability within the amphibious objective 

area. On the morning of 15 June 1944, two reinforced Marine Corps divisions, the Second and 

Fourth, stormed ashore over four beaches in the southwest portion of the island. Resistance was 

fierce. The Army’s 27th Infantry Division, acting as the floating reserve, was committed to the 

fray to support the Marine attack and remove them from harm’s way once it was learned the 

Japanese First Mobile Fleet had sortied.  

Despite being largely unprepared for this contingency, Vice Admiral (VADM) Spruance 

and the Fifth Fleet reacted swiftly in anticipation of a climactic battle at sea. A detachment of 

Patrol Squadron (VP) 16’s new-to-theater PBM-3D Mariner patrol planes was ordered to Saipan 

and anchored on the congested Saipan Shelf in an open ocean seadrome to provide long-range 

Radio Direction and Ranging (RADAR) surveillance for the fleet. A Mariner from this 

detachment detected the Japanese fleet at extended range, but a communication problem 

thwarted their efforts to provide a contact report until after their return. The U.S. Navy lost the 

initiative but, thanks in part to the late report, was prepared for the coming aerial assault.  

The Fifth Fleet absorbed wave after wave of Japanese aerial attacks but broke the back of 

the Japanese naval air arm during the ensuing “Great Marianas Turkey Shoot.” A risky evening 

U.S. maximum-range carrier counterattack, the “Mission Beyond Darkness,” severely damaged 

the Japanese fleet. Still, prudence demanded that the carriers remain near the invasion fleet to 

protect them adequately, so aggressive pursuit was not an option. The Battle of the Philippine 

Sea ended as the bloodied but intact Japanese fleet escaped to the west.   



   

 

6 
 

Ashore, combat continued to rage unabated. Progress was painfully slow, but the Marines 

and soldiers made steady gains directly enabled by overwhelming U.S. firepower. They cleared 

most of the island and pushed Japanese resistance into a pocket of defense in the northern portion 

of the island. Then, in the early morning darkness of 7 July 1944, the remnants of the Japanese 

garrison committed to a final full-scale bonsai charge known as a gyokusai (Cabrera 2015:21–

22). Despite a spirited defense, the gyokusai quickly overran two Army 27th Infantry Division 

battalions. This attack culminated in front of the tubes of the 10th Marine’s 155mm artillery 

pieces and collapsed. Although organized Japanese resistance on the island ended, mopping-up 

operations continued for many months. The Battle of Saipan was over.  

That Saipan is not more prominently known is likely due to the timing of the invasion. 

Just nine days before the Marines stormed ashore, Operation OVERLORD, the long-anticipated 

Allied invasion of Europe, commenced. All eyes were on the decisive battle shaping the war 

against Germany. This does not diminish the criticality of the conquest of the Marianas in any 

way. After the Battle of Saipan ended, the U.S. had its coveted airfield, and B-29 Superfortresses 

began striking Japan from Isley Field (formerly Japan’s Aslito Field) in November 1944 (Hallas 

2019:458). The neighboring island of Tinian played host to the B-29s that delivered the first — 

and only — nuclear weapons ever used in war. As the U.S. celebrated its victory in the Marianas, 

the Japanese lamented defeat and recognized the full scope of their loss. Premier Tojo remarked, 

“Japan has come to face an unprecedently great national crisis” that ultimately toppled his 

military faction. Yet Fleet Admiral Osami Nagano best summarized Japanese sentiment when he 

said of the loss of the Marianas, “Hell is upon us” (Crowl 1960:1).  

 

 



   

 

7 
 

1.2. PREVIOUS ARCHAEOLOGICAL EFFORT 

 Given the historical significance of Operation FORAGER, especially the Battle of Saipan 

component, there has been no shortage of recent archaeological efforts dedicated to 

rediscovering, documenting, and interpreting the past. This includes numerous archaeological 

surveys beginning in the 1980s through today. These projects ranged from National Park Service 

(NPS) surveys, surveys directed at master’s theses, Commonwealth of the Northern Marianas 

(CNMI) Historic Preservation Office (HPO) efforts, and others (Thomas and Price 1980; 

Miculka and Manibusan 1983; Miculka et al. 1984; Carrell 1991; Pacific Basin Environmental 

Consultants (PBEC) 1985; Burns 2008a; Burns 2008b; Guerrero 1987; Bell 2010; McKinnon 

and Carrell 2011; DeFant 2012; McKinnon and Carrell 2014; Pruitt 2015; and Adamson 2020).  

Due to the submerged lands mandates of the NPS, HPO priorities on heritage tourism, 

and the prolific quantities of archaeological and culturally significant materials inshore, previous 

efforts focused almost exclusively on Tanapag and Garapan Lagoons near Tanapag Harbor and 

Flores Point seaplane base areas. Consequently, little effort has been dedicated to locations 

beyond the reef and deeper water, such as the Saipan Shelf. Even as the near-shore 

archaeological site surveys continue, attention may shift to places beyond the barrier reef. As 

such, this thesis turns its attention west across Saipan’s barrier reef and makes an inaugural 

archaeological effort in deeper water.  

1.3. LITERATURE REVIEW 

A historiography and literature review of the immense body of work addressing the 

Pacific War reveals extraordinarily little contributory material regarding the employment of 

seaplanes during Operation FORAGER. Books addressing the war's operational and strategic 

levels are almost voiceless concerning long-range seaplane searches during the Battle of the 



   

 

8 
 

Philippine Sea. Alternately, they extensively address concurrent search efforts by the established 

submarine screen, Task Force 58’s short-range carrier search efforts, and even the intelligence 

community’s efforts to intercept, decipher, and direction find Japanese radio transmissions. 

Biographies of the key U.S. leaders during the battle, which one could expect to address 

all the resources these leaders employed broadly, and efforts made to find, fix, and destroy the 

Japanese fleet, are similarly silent on the issue. VADM George Carroll Dyer wrote the two-

volume biography on VADM Richmond Kelly Turner, Commander, Joint Expeditionary Force, 

Task Force 51, including his actions during Operation FORAGER, at the behest of Rear Admiral 

(RADM) Ernest “Judge” Eller, then acting as Director of Naval History, Naval History Division 

(Dyer 1972). The Amphibians Came to Conquer does not mention the activities of seaplanes or 

seaplane tenders during the Saipan campaign, even though VADM Turner was personally 

responsible for ensuring that USS Ballard (AVD-10), a destroyer seaplane tender that provided 

initial support for seaplane operations at Saipan, was initially present with the invasion fleet. 

E.B. Potter’s (1976) definitive biography of Admiral Chester Nimitz, eponymously titled Nimitz, 

also speaks extensively regarding the contributions of submarines, carrier aircraft, and 

intelligence efforts but capriciously omits any reference to PBM Mariner activities or 

observations.  

Thomas Buell’s (1974) biography, The Quiet Warrior, about VADM Spruance, the man 

responsible for the conduct of the Fifth Fleet during the Battle of the Philippine Sea, does a more 

thorough, yet still subdued, job of addressing the participation of VP-16 PBMs during this 

campaign. Buell speaks to Admiral Spruance’s direction to VADM John Hoover, Commander, 

Central Pacific Forward Area, and Pacific Ocean Area, to deploy long-range seaplanes to Saipan 

in response to Flying Fish’s detection of the Japanese fleet passing through San Bernadino 



   

 

9 
 

Straits. After a follow-on sighting by submarine Sea Horse, Buell states, “Hoover was to send the 

long-range seaplanes to Saipan to begin immediate searches to the west” (Buell 1974:263). A 

third and final mention of long-range seaplanes communicated VADM Spruance’s annoyance at 

receiving the VP-16 sighting report eight hours late.  

Ronald Spector (1985), well regarded for his definitive single-volume account of the 

Pacific War titled Eagle Against the Sun, commits only a single sentence: “A contact report by a 

long-range flying boat which had found the Japanese fleet an hour after midnight never reached 

Task Force 58” to the subject (Spector 1985:309). This statement is misleading. The contact 

report reached VADM Spruance, albeit eight hours late, but still in time to provide a warning that 

helped the U.S. fleet anticipate and prepare for the coming Japanese onslaught.  

In a later book, At War at Sea, Spector (2002) omits any reference to the seaplanes’ 

participation in the Battle of the Philippine Sea. More egregious is his statement that no air 

search occurred during the night of 20 June 1944. Following the Great Marianas Turkey Shoot 

and the Mission Beyond Darkness, the U.S. admirals allowed their Task Force 58 pilots much-

needed rest following a day of particularly grueling combat. While they slept, six RADAR-

equipped PBM-3D Mariners were again prowling the vast Pacific expanses, searching for the 

enemy fleet to allow morning strike targets for the rested carrier pilots. Although those searches 

were unsuccessful, Spector again inadvertently leaves the reader with a false impression that 

materially misconstrues the conduct of the battle.  

In his definitive book Combined Fleet Decoded, John Prados (1995) focused on the U.S. 

intelligence efforts against the Japanese. He addressed radio intercepts and direction-finding 

activities during the Battle of the Philippine Sea. He singles out naval vessels, saying “American 

submarines complemented the radio intelligence units” as supporting these intelligence activities 



   

 

10 
 

(Prados 1995:570). He also communicated that the Japanese, in their eastward transit, made sure 

to remain clear of the search radius of long-range B-24 Liberators based in Manus in the 

Admiralty islands (one of which, a B-24 from the 371st Bomber Squadron, 307th Heavy 

Bombardment Group, 13th Air Force was piloted by Lieutenant Anthony). Yet, in a glaring 

omission, he inexplicably makes no mention of the detachment of PBM-3D Mariners that moved 

forward to Saipan expressly to fill the gap in long-range aerial search requirements. 

Volumes that address the Japanese perspective of the war and dedicate copious pages 

specifically to the Battle of the Philippine Sea as a pivotal engagement are similarly silent on the 

issue of U.S. long-range patrol plane capabilities and operations. John Toland’s (1970) The 

Rising Sun, and especially Paul Dull’s (1978) A Battle History of the Japanese Navy, speak to 

Japanese battle plans that relied upon their longer-range carrier aircraft and ability to operate 

from nearby bases such as Truk, Yap, Guam, and Rota to provide them with the naval search 

advantage with which to seize the operational initiative. Attention is given to Japanese 

considerations regarding the risk of enemy submarine or carrier-based search aircraft detection. 

Yet, their planning calculus seemingly never addressed the potentiality of U.S. long-range 

seaplane search aircraft. Likely, they felt confident their adversaries could not forward deploy 

long-range search aircraft to an unprepared expeditionary location during the height of combat 

operations. If so, their confidence was misplaced.  

Samuel Eliot Morison (1953) stands alone amongst the preeminent authors of the Pacific 

War in his thorough treatment of the long-range seaplane participation and contribution to the 

Battle of the Philippine Sea. In Volume VIII, New Guinea and the Marianas, of his History of 

United States Naval Operations in World War II series, he details the decision to deploy PBM 

aircraft from Eniwetok to the open ocean seadrome off Saipan (Morison 1953). He goes on to 



   

 

11 
 

address the extreme operational challenges these aircraft endured while executing their mission 

from their atypical anchorage and friendly fire hazards faced by allies unaccustomed to 

seaplanes, especially the new arrival and unfamiliar PBM Mariner, operating in the forward area 

during an amphibious assault without a prepared seadrome. Most importantly, Morison goes into 

detail in addressing the potential campaign-changing detection of the Japanese fleet and 

subsequent communications failure by one of these aircraft on the eve of the battle. No other text 

speaks to the operational challenges regarding the deployment or provides this context, as does 

Morison’s (1953). 

Other volumes dedicated to the operational and tactical execution of the battle fare no 

better in their attention to the PBM issue. In his book Battles of the Philippine Sea, VADM 

Charles Lockwood (1967), who served as Commander, Submarines, Pacific Fleet, during the 

war, addresses only the unreceived contact report provided by PBM pilot Lieutenant (LT) Arle, 

absent amplifying information.  

In Red Sun Setting (1981), William Y’Blood does a better job of addressing the searches 

conducted by the PBMs on the night of 17–18 June 1944 and again focuses extensively on the 

missed contact report on the night of 18–19 June 1944. Interestingly, on this book’s end sheet art 

is a Task Group 58.1 Combat Information Center chart that depicts a friendly fire engagement in 

which a division of F6F Hellcats attacked a returning PMB just west of Guam. The chart entry 

reads, “Coke 1 attacked PBM 2V2U by mistake and heavily damaged it” (Y’Blood 1981:end 

sheet). This engagement played a critical role in creating the conditions for the subsequent loss 

of the aircraft central to this thesis. However, no other Task Force 58 information regarding the 

incident has yet been located. 



   

 

12 
 

The passage of time, enhanced availability of primary source documents, and previous 

academic work on the subject have not positively impacted this imbalanced perspective. Evan 

Mawdsley’s (2024) Supremacy at Sea, dedicated to Task Force 58 operations, lacks information 

regarding seaplane activities in the Marianas despite devoting 71 pages to operations there.  

Other books that aim to address the air war in the Pacific holistically are also scant on 

details regarding the services provided by seaplanes. Barrett Tillman’s (2011) Whirlwind: The 

Air War Against Japan 1942-1945, a decidedly AAF-centric book, devotes a scant 4 of 336 pages 

to seaplane operations; these exclusively address their activities in terms of their ability to rescue 

the aircrew of fighters and bombers lost in combat. Eric Bergerud’s (2000) Fire in the Sky: The 

Air War in the South Pacific is also painfully short on information regarding seaplane operations, 

never once so much as mentioning the PBM Mariner, an indispensable component of the Navy’s 

seaplane fleet, despite being a 723-page tome. 

Although hundreds of books are devoted to aviation operations during World War II, just 

three are dedicated to holistic seaplane operations during this conflict. Michael Kammen penned 

two of these at the request of the Navy’s Research Division of the Bureau of Aeronautics. These 

are the paired Operational History of the Flying Boat, one addressing selected campaigns in the 

Pacific (1959) and the other European operations (1960). The last title considered is David 

Bruhn’s (2019) Eyes of the Fleet: The U.S. Navy’s Seaplane Tenders and Patrol Aircraft of World 

War II. This relative newcomer comprehensively accounts for Pacific theater-wide seaplane 

operations.  

There is a glaring gap in the literature associated with World War II seaplane operations 

and the employment of VP-16 PBM-3D Mariners during the Battle of the Philippine Sea. 

Beyond the general omissions of their presence, this inattention of the subject sows a mistaken 



   

 

13 
 

perception of when and how they were employed. Indeed, the almost singular focus on the failed 

contact report gives the impression that not only were these aircraft ineffective in a search role, 

but even when successful, their crews could not convey this critical information properly. Except 

in two notable cases, Morison’s (1953) New Guinea and the Marianas and Y’Blood’s (1981) Red 

Sun Setting, there is a complete lack of context regarding the unplanned immediacy of the 

deployment, the insufficient seaplane tender support, or the extraordinarily harsh operating 

environment these aircraft were thrust into. In this case, context is everything. Only by closing 

these gaps in the existing literature can the actual contributions of VP-16’s deployment to Saipan 

be understood and properly appreciated.  

1.4. RESEARCH AIMS AND QUESTIONS 

This thesis strives to help close gaps within the existing literature by answering one 

primary and four supporting research questions: 

1. What conditions and characteristics led to the innovative establishment of the open 

ocean seadrome, and what contributions did it and the aircraft it supported have to the 

Battle of the Philippine Sea and the overall war effort? 

1.1.How effective was the seaplane tender support in the open ocean seadrome, and how 

did it influence future seaplane tender operations? 

1.2.What archaeological remains, if any, of Saipan’s original open ocean seadrome still 

exist? 

1.3.What can the archaeological investigation of the bowser boat, anchorages, and PBM 

Mariner Bureau Number (BuNo) 48173 communicate regarding the operations within 

the open ocean seadrome? 

1.4.How can site formation studies of the BuNo 48173 deep-water wreck contribute to 

understanding World War II seaplane operations? 



   

 

14 
 

Addressing these questions will explore the underrepresented contribution of seaplanes 

and seaplane tenders during the opening stages of Operation FORAGER. It will set the 

conditions to locate and archaeologically survey what remains of an atypical seadrome explicitly 

established to support this operation and how it was operated and optimally lead to the discovery 

of the wreckage and remains of a PBM-3D Mariner, a refueling bowser boat, and residual 

anchorage. 

These artifacts represent an enduring archaeological legacy created in just five days 

during June 1944. Finding them will enhance and contextualize a deeper understanding of the 

Battle of the Philippine Sea and the war material involved in this event. Due to the unique 

circumstances surrounding the loss of this aircraft, it will also have an outsized impact on our 

understanding of site formation theory. The use of innovative new technology in searching for 

this site is also important for the continued evolution of maritime archaeology as a discipline.  

Equally important is evaluating the impact of the open ocean seadrome on the historical 

record. Although exceptionally well documented, the Battle of the Philippine Sea remains highly 

controversial due to some decisions made by some United States commanders during the battle, 

namely in not being more aggressive in prosecuting attacks on the Japanese and willfully ceding 

the initiative to them. Understanding the contribution, or lack thereof, by the patrol planes 

operating from the open ocean seadrome materially contributed to the mindset and decision-

making by these commanders. This misunderstood aspect of the battle and its impact on its 

conduct shed a brighter light on this still controversial subject. Achieving this thesis’s goals will 

enhance and contextualize this culturally significant site's archaeological and historical legacy.  

 

 



   

 

15 
 

1.5. HISTORICAL AND ARCHIVAL RESEARCH 

Given the lack of secondary source information regarding the experience of VP-16 and 

their PBM-3Ds during Operation FORAGER, as discussed in the literature review, a determined 

effort to research primary sources was required. Much of this could be accomplished online via 

the digital military record database Fold3. The National Archives have done an exceptional job 

digitizing and making available to the public the prolific primary source documents, many of 

which were relied upon heavily while conducting research for this thesis. Primary source 

documents were prioritized, although secondary sources added context or corrected the record 

based on new information.  

Chief among these were unit diaries and histories of U.S. squadrons, most notably VP-16, 

whose sunken Mariner is the focal point of this thesis, and supporting vessels and commands. 

Operational and attack orders regarding Operation FORAGER writ large, as well as those 

specific to the invasion of Saipan (Operation TEARSAIL), provided invaluable information 

governing the planning and conduct of the invasion. Similarly, after action reports were 

extensively reviewed. The National Archives is a trove of quality images from the Battle of 

Saipan, and many found their way to these pages, which visually reinforced some of the 

discussed concepts. Finally, a review of captured Japanese intercepts and intelligence rounded 

out this effort. A mosaic of information emerged from these documents that, when combined, 

formed a sum greater than its component parts.  

This thesis's units of measurement use specific metrics employed in these primary 

sources and associated metric/imperial conversions as required. Although this occasionally 

presents an inconsistent methodology in the text, it is appropriate to maintain faith to the primary 

sources.  



   

 

16 
 

Unfortunately, in some cases, documents associated with the campaign have yet to be 

declassified and, therefore, are unavailable. In 2009, President Barack Obama signed Executive 

Order 13526, which established a standardized protocol for declassifying documents with 

permanent historical value after the 25-year mark. It has not been evenly applied. Naturally, 

government agencies remain committed to having proper declassification authority review 

documents before making them available. There is a shortfall in the capacity to do so, which 

invariably creates gaps in document availability.  

The situation regarding VADM Hoover’s personal and professional papers, fourteen 

boxes held at the Naval War College, is instructive. Because he was responsible for the 

operations of all air bases and patrol plane activities in the Pacific Theater, including during 

Operation FORAGER, this collection undoubtedly contains germane material that remains 

classified and unavailable that could materially inform this thesis. Hoover’s papers have not yet 

been declassified, inhibiting their contents from informing the historical record and future 

potential archaeological efforts. 

If there were a singular failure in primary source research, it was the inability to locate 

the Fifth Fleet’s anchoring plan for the invasion of Saipan. This single document, a ubiquitously 

distributed map provided in duplicate to every ship involved in the Saipan invasion, conclusively 

shows the anchoring locations managed by the Senior Officer Present Afloat (SOPA). Most 

vessels in their unit diary specify which anchoring berth they were in when specific incidents or 

activities occurred. Locating this map would open the aperture on numerous aircraft and surface 

vessel loss locations often provided as a relative bearing and distance from a specific anchoring 

berth. Obtaining this document could dramatically shrink the proverbial haystack and enable the 



   

 

17 
 

discovery of many known lost needles, the locations of which are often referenced by direction 

and distance from anchoring berths. 

1.6. TERMINOLOGY 

Finally, a brief note on the open ocean seadrome terminology used throughout this paper 

is warranted. Typically, seadromes were established in sheltered lagoons or harbors to shield 

aircraft from harsh environmental factors, such as ocean swell, and protection from prevailing 

winds. These were accompanied by a robust infrastructure ashore that provided maintenance 

facilities, arming and fueling points, barracks and life support facilities for crew and maintainers, 

and even hangars to protect aircraft from the elements completely; significant quantities of fuel, 

ammunition, and other material support were stored there. Dedicated runaways within this 

protected area that had been carefully swept of obstacles, established outside regular maritime 

traffic lanes, and often marked with lights or buoys were common. 

However, it was accepted that seaplanes might sometimes be compelled to use advanced 

basing, borne of operational necessity when required. The 1942 Current Tactical Orders and 

Doctrine, Volume 3: Patrol Aircraft, prescriptively laid out requirements in these cases. An 

advanced base’s paramount consideration should be safeguarding plane operations from attacks 

by enemies due to their vulnerability. They were recommended to be “close to the area of active 

operations” yet “desirable that they be sufficiently remote from enemy bases to decrease the 

probability of their being seriously damaged by enemy air raids” (Patrol Aircraft 1942:73).  

Ammunition, fuel, and personnel should be dispersed in isolated areas ashore. Similarly, 

seaplanes and their support vessels were to use inlets and bays and leverage headlands and 

islands to obscure their presence from the enemy. If detected, the manual recommended they be 

immediately relocated to a discrete location, even at the expense of scouting operations. 



   

 

18 
 

Concealment was the primary defense against destruction, as well as scattering the moorings to 

properly disperse the aircraft and using dummy aircraft to distract any attacker. An air attack was 

considered the most significant threat, but enemy surface and submarine actions were not ruled 

out. Land-based RADAR equipment with shore and carrier-based fighter support was essential to 

defending an advanced seaplane base (Patrol Aircraft 1942:73). 

Advance bases were established in “roadstead” environments to protect from the 

elements. Roadstead anchorages are “semi-protected offshore anchorages for oceangoing 

vessels” (Port Construction 1964:22). A roadstead uses reef systems or islands to protect 

seaplanes from damaging ocean swells and provide a breaker for high winds. Some level of 

shore facilities was still necessary for the effective operation of patrol seaplanes, along with 

sufficient fuel and material support. Although a roadstead anchorage enjoying these 

characteristics was suboptimal compared to a fixed base with advanced and specialized 

infrastructure, it was enough to support seaplane operations on an enduring basis with these 

features.  

The initial seaplane anchorage at Saipan had no roadstead characteristics. It was well 

outside the island’s barrier reef in the open ocean. Established 4 mi. (6.5 km) from the island, it 

offered no protection from ocean swell and very little from prevailing winds from the east. It had 

no access to shore facilities, which at that time were still under complete control of the Japanese 

defenders.  

In addition to attack by enemy aircraft as anticipated by Current Tactical Orders and 

Doctrine, Volume 3: Patrol Aircraft (1942), VP-16 PBMs found themselves under constant 

observation by the Japanese on the island in defiance of doctrinal advance base guidelines. They 

were also sometimes taken under fire by enemy shore batteries, a situation nobody could have 



   

 

19 
 

conceived of, even in the days immediately leading up to the invasion of Saipan. None of the 

characteristics associated with a roadstead seadrome were available to the squadron, save a 

sufficiently shallow anchoring point on the Saipan Shelf to affix themselves, established only the 

night before their arrival. In sum, their initial anchorage was not a roadstead environment by any 

metric.  

For this reason, this thesis uses the non-traditional term “open ocean seadrome” to reflect 

the unique operating conditions of this anchorage. Although numerous sources holistically refer 

to the location as the “Garapan anchorage,” “Garapan Roads,” or “Saipan Seadrome,” none of 

these were specific to the initial open sea operating area assigned to the VP-16 PBMs (Adamson 

2020:36; Kammen 1958:8). Later in the campaign, as the situation ashore improved, these 

aircraft continued to shift forward toward a proper roadstead anchorage and eventually occupied 

the repurposed permanent Japanese seaplane base in Tanapag Harbor. However, the activities 

addressed in this thesis took place exclusively in the first five days of the campaign from 17–21 

June 1944, well before this landward march began and completely within the confines of the 

open ocean seadrome. 



 

CHAPTER 2: HISTORY 

 To fully appreciate the archaeological and historical legacy of the open ocean seadrome, a 

foundational knowledge of the U.S. Navy’s operational objectives in establishing this innovative 

facility must be understood. Similarly, a knowledge of the men and machines, seaplane and 

seaplane tender alike, leveraged to achieve these goals, is instrumental. 

2.1. PBM-3D MARINER 

 

The U.S. Navy, seeking to professionalize and modernize during the 1930s, devoted 

considerable attention to the aviation patrol arm that served as the fleet's probing eyes. This 

included the seaplanes and tenders that cared for them and developed a new generation of pilots. 

Understanding the development and addition of the PBM and subsequent evolution of the PBM-

3D to the patrol air arm and the criticality of this airframe is essential.  

The Navy incorporated the instantly recognizable Consolidated PBY Catalina as a “patrol 

bomber” in 1935. Attention soon turned to complementing this aircraft with a larger, heavily 

armed flying boat capable of greater range and enhanced offensive capability. Consolidated, 

Sikorsky, and the Glen L. Martin Company submitted designs for a four-engine seaplane. 

Although Consolidated won the contest with a design that eventually became the PB2Y 

Coronado, the Martin Company continued its design using company funds. The Navy, realizing 

the excessive cost of purchasing and operating the new four-engine Coronado, opened another 

competition for a two-engine variant aircraft to bridge the gap between the less capable PBY 

Catalina and the exorbitantly expensive PB2Y Coronado (Hoffman 2004:1).  

The Martin Company’s entry, Model 162, provided enhanced speed, load, and armament 

advantages compared to the PBY Catalina while costing significantly less than the PB2Y 

Coronado. This entry was the winning combination the Navy was looking for, and the Martin 



   

 

21 
 

Company earned a contract to field a single prototype dubbed the XPBM-1; XPB being the 

Navy’s terminology for Experimental Patrol Bomber and the M represented “Martin.” The 

Model 162A “Tadpole Clipper” was a successful 3/8s scale model; its performance validation 

expedited the procurement of the entire contract, and the Navy offered Martin a production 

contract in1937. Soon, 21 PBM-1s were delivered for $5.1 million. The original XPBM-1 

Bureau Number BuNo 1939 took to the air on 13 February 1939 (Hoffman 2004:2–4; 12–14; 

Wings 1972:8).  

The Navy placed a follow-on order for 379 PBM-3s in November 1940. Significant 

improvements were made despite an outwardly similar appearance to its predecessor. Additional 

armor was incorporated, engine nacelles were lengthened to double as the airplane’s bomb bay to 

accommodate an increased ordnance payload, and fixed-wing floats replaced the retractable 

version. A shift to enhanced Wright 2600-12 1,700 horsepower engines, each providing an 

additional 100 horsepower over the previous version used by the PBM-1, was the most crucial 

change (FIGURE 2.1). Engine reliability and power problems plagued the PBM-3, even with the 

newer engine and extra horsepower (Hoffman 2004:6).  

Mariner variants continued to evolve to meet the unique needs of various theaters and 

mission sets after war descended upon the nation. A PBM-3S antisubmarine version, PBM-3R 

transport, and a PBM-3D patrol bomber became widely employed, while a single stand-alone 

PBM-3E RADAR testbed was also developed. In all, 677 PBM-3s were eventually constructed 

to meet Navy requirements (Hoffman 2004:9).  

The first PBM-3 variant to arrive for operational use was the PBM-3R transport, which 

fulfilled a desperately needed National Air Transport Service (NATS) requirement. These aircraft 

had reinforced flooring, cargo doors, and hoists to facilitate their cargo-centric roles and handle 



   

 

22 
 

their 8,000-pound (lbs.) (3,629 kilogram [kg]) capacity load, including up to 20 people (Ginter 

2013:34; Hoffman 2004:18–19). 

The PBM-3C Patrol Bomber was introduced by the Navy as an upgraded version of the 

baseline PBM-3 which added self-sealing fuel tanks, additional guns, and a search RADAR for 

detecting submarines (Ginter 2013:37). On 1 December 1942, the first PBM-3Cs were assigned 

FIGURE 2.1: Wright 2600-14 engine displayed at the Pima Air Museum, Arizona (Image courtesy of Pima Air 

Museum). 



   

 

23 
 

to Patrol Squadron 32 for service in the Panama Canal Zone gradually spreading throughout the 

Atlantic Theater to squadrons assigned anti-submarine warfare (ASW) duty (Hoffman 2004:19). 

The Navy recognized the ASW campaign in the Atlantic required further theater specific 

optimization of the PBM-3C. These modifications capitalized on the near absence of enemy 

airborne opposition and optimized for the ranges needed. On 13 July 1943, the Navy held an 

Anti-Submarine Measures Conference that updated the evolution of a stripped-down variant that 

removed 6,400 lbs. (2,903 kg) of equipment, creating the PBM-3S (Anti-Submarine Conference 

1943:2): 

The items being removed consist of the following major ones bow, upper and tail turrets, one side waist 

gun, all armor, bomb-sight and stabilizer, self-sealing fuel cells, oxygen, large auxiliary power plant, most 

comfort items including galley table and seats, all but two bunks, all lockers, galley and sink, including 

cabinets and refrigerator, smoke tank installation, one of engine work platforms, hoisting sling and a 

number of smaller items. (Anti-Submarine Conference 1943:2)  

 

 The distinct operating conditions of the Pacific Ocean demanded yet another variant of 

the PBM-3. Unlike their Atlantic cousins, Mariners assigned to the Pacific performed myriad 

duties, just one of which was ASW patrols. They ranged ahead of the U.S. fleets during the 

island-hopping campaign, scouring the ocean for signs of Japanese air, surface, and submarine 

activity. The Imperial Japanese Navy (IJN) retained a capable fleet that posed a genuine threat to 

the U.S. advance until it was finally crippled during the Battle of Leyte Gulf in October 1944. 

Still, IJN elements, including the super battleship (BB) Yamato, sortied from the home islands 

during the Okinawa campaign in 1945 and might have inflicted damage on the invasion fleet if 

not for the early detection by PBM Mariners.  

Mariners probed the vast distances of the Pacific, searching for the enemy, and often 

encountered enemy patrol aircraft doing the same. Thanks to strong defensive armament and 

robust armor, the PBM-3D was generally advantaged when these encounters developed into 

running gunfights between patrol aircraft. Yet such aerial engagements paled compared to 



   

 

24 
 

encounters with Japanese interceptors and combat air patrols (CAP). Mariners routinely attacked 

targets protected by an aegis of Japanese Mitsubishi A6M Zeros, Nakajima A6M2-N Rufe 

floatplanes, or other fighter variant aircraft purpose-built to shoot enemy aircraft down. PBMs 

often faced heavy anti-aircraft fire while attacking Japanese shipping or shore installations. Some 

conducted search and rescue patrols to locate and recover survivors of shot-down or otherwise 

disabled aircraft and sunken ships. These rescues sometimes necessitated a harrowing landing in 

the open ocean, risking one’s aircraft and life to save another.  

A sturdy, heavily armed, and armored patrol seaplane was required to enter this fire 

cauldron and emerge safely on the other end — it was the PBM-3D Mariner, optimized for 

combat conditions in the Pacific (FIGURE 2.2). It possessed self-sealing fuel tanks, three power 

turrets, a Norden bombsight, and enhanced armor protection for the crew and critical aircraft 

components. It was upgraded with 1,900-horsepower Wright R-2600-22 engines and new four-

bladed propellers to assist with the power requirements of this added weight. These, too, proved 

underpowered for the challenging operations that awaited the aircraft despite being a marked 

improvement over the original PBM-1’s Wright 2600-6 1,600-horsepower and the earlier PBM-3 

variant’s Wright 2600-12 1,700-horsepower engines.  



   

 

25 
 

 

FIGURE 2.2: Schematic of the PBM-3D (Image from Handbook of Flight Operating Instructions for 

U.S. Navy Model PBM-3D). 



   

 

26 
 

An integrated search RADAR provided day and, importantly, night search capability and 

was a significant shared feature with the PBM-3C/S. The AN/APS-15 “Mickey” air-to-ground 

RADAR system was placed in a prominently covered dome on top of the aircraft just aft of the 

cockpit. Its operator sat just aft of the flight deck (Ginter 2013:42). This system provided a 

capable sensor to conduct night surface searches in the relative security of a blackened sky to 

protect this still vulnerable patrol plane from the enemy. VP-202 operated this RADAR in 

Saipan, reporting reliable detection of land formations up to 90 mi. (145 km) away and surface 

ships between 40 and 50 mi. (64 and 80 km) distance, well beyond visual detection ranges during 

daytime operations (VP-202 1944:4). This system gave the U.S. a significant edge in maritime 

patrol, allowing long-range searches conducted around the clock, day and especially night. 

The aircraft’s Handbook of Flight Operating Instructions for U.S. Navy Model PBM-3D 

(1944) described it as “a two-engine high-wing, medium-range flying boat which may be used as 

a patrol airplane, bomber, or torpedo airplane. The approximate dimensions are listed as: length, 

79 feet 10 inches; height, 28 feet 2 inches; and span, 118 feet. The normal gross weight is 46,500 

pounds” (FIGURE 2.3). Significantly, its fuselage was divided into five watertight compartments 

like those of a ship. The aircraft maintained positive buoyancy in the event of a single-

compartment flood. Unfortunately, this was the case if two adjacent watertight compartments 

were compromised, as dramatically demonstrated during the Battle of Saipan. These aircraft 

played a dominant role in the patrol arm and campaigns in the Pacific. When they arrived as 

reinforcements, they were a welcome addition to the force (FIGURE 2.4). 



   

 

27 
 

 

FIGURE 2.3: Top: Three-view drawing for PBM-3D. Bottom: Three-quarter view of PBM-3D (Image 

from Handbook of Flight Operating Instructions for U.S. Navy Model PBM-3D). 
 



   

 

28 
 

 

FIGURE 2.4: 1942 World War II Martin Aircraft propaganda poster featuring the Martin PBM Mariner (Image 

courtesy of the Glenn L. Martin Company). 



   

 

29 
 

2.2. PATROL SQUADRON 16 (VP-16) 

 

VP-16 stood up at Naval Auxiliary Air Station (NAAS) Harvey Point, North Carolina, on 

20 December 1943. In addition to the aircraft, the Martin PBM-3D Mariner, the squadron 

received its officers and sailors, commanded by Lieutenant Commander (LCDR) William J. 

Scarpino (FIGURE 2.5). As with any massive military buildup, these personnel were largely 

untested and extremely limited in operational experience. Only 4 lieutenants and 4 lieutenants 

(junior grade) out of 61 officers had any operational or combat experience beyond the squadron 

commander. All the other pilots reported directly from Naval Air Station (NAS) Banana River, 

Florida, following two months of initial training in PBMs (Patrol Squadron 16 War History [VP-

16 History] 1945:8,12). Although they all had basic ratings in the PBM, few in the squadron had 

ever seen, much less flown, the new PBM-3D variant before arriving at NAAS Harvey Point. A 

parallel challenge existed within the squadron ground officer experience (VP-16 History 

1945:Annex 3). A robust squadron program sought to improve this inexperience.  

The next four months were spent at NAAS Harvey Point conducting extensive training 

activities, heavily emphasizing ASW operations in anticipation of this likely mission. A week of 

advanced training was performed later at NAS Key West, using friendly submarines as active 

targets for squadron searches and mock attack runs, and introduced the squadron to the use of 

sonobuoys (VP-16 History 1945:16–17). Although the NAS Key West ASW training was 

excellent by all counts, it was suboptimal in preparing the aircrews for the long-range night 

RADAR searches they ultimately conducted in the Marianas. Also, due to the compressed 

timeline and scarcity of seaplane tenders, training for operations from these seaborne platforms 

was necessarily omitted. Finally, adverse weather, engine problems, and material support 

shortfalls conspired to hinder the execution of the total training syllabus dramatically. These 



   

 

30 
 

training limitations later hampered the success of the squadron’s first deployment (VP-16 

History 1945:12).  

Less than five months after commissioning, VP-16 was sent to the Pacific to assume 

combat duties. After a challenging transit from the U.S. plagued by engine problems, resulting in 

three airplanes and one aircrew crew lost, the squadron was fully assembled and settled into 

Eniwetok by 13 June 1944 (VP-16 History 1945:32–34). Just three days later, on the evening of 

FIGURE 2.5: LCDR William J. Scarpino on 13 February 1945 during the time he was in command of VP-16 (Photo 

courtesy National Archives photo #80-G-302684). 



   

 

31 
 

16 June 1944, the officers were called together for a brief at 2130. By midnight, all hands knew 

the squadron was to deploy to the forward combat zone — destination Saipan immediately. This 

compressed deployment timeline did not afford the luxury of a proper briefing, even if 

information or intelligence was available.  

VP-16 aircrews took off only three days after they arrived in the forward theater and a 

mere eight hours after learning of their destination. With no understanding of their mission or 

role in Saipan, they departed, not knowing where they were to land, anchor, or be supported. 

Such was the situation's urgency and the need to get long-range RADAR-equipped PBMs “on 

station” in Saipan (FIGURE 2.6). Six crews and two backups were designated for a 0600 launch 

on 17 June 1944, led into Saipan by their commanding officer, LCDR Scarpino (VP-16 History 

1945:35). 

2.3. SEAPLANE TENDERS 

The capability to establish and maintain the open ocean seadrome at Saipan was entirely 

predicated on the ability of seaplane tenders to set and sustain moorings and tend to the Mariners' 

aircraft and crews. Three seaplane tenders participated: USS Ballard (AVD-10), USS Onslow 

(AVP-48), and USS Pokomoke (AV-9). Each was a different class vessel with wide-ranging 

FIGURE 2.6: VP-16 PBM-3D Mariner as configured in September 1944, shortly after Operation FORAGER (Image 

courtesy Wings Aviation). 



   

 

32 
 

capabilities, but all contributed to operations in the open ocean seadrome (Bruhn 2016:314). It is 

necessary to understand the capabilities and limitations of the participating seaplane tenders to 

appreciate the services rendered to VP-16’s operations from 16–21 June 1944. 

2.3.a. USS Ballard (AVD-10) 

USS Ballard was the most consequential tender for the initial operations of the open 

ocean seadrome. It began as one of 156 Clemson-class four-stack destroyers produced during the 

U.S.’s massive World War I wartime building effort. Midshipman Edward J. Ballard, killed 

during the storied engagement between USS Chesapeake and HMS Shannon on 1 June 1813 

during the War of 1812, was the vessel's namesake. Ballard was completed after the drums of 

World War I fell silent, and too late to participate in that conflict. Bethlehem Steel Corporation 

launched the vessel in Squantum, Massachusetts, on 7 December 1918. It was commissioned on 

5 June 1919. Initially assigned to the Atlantic Fleet, Ballard led an undistinguished existence 

cruising throughout Europe and the Mediterranean (FIGURE 2.7). It was later transferred to the 

Pacific Fleet, decommissioned, and placed in reserve on 17 June 1922, just three years after its 

original commissioning (Mooney 1959:88). 

Ballard was deemed too ancient and vulnerable for frontline service in a contemporary 

war, yet it retained value for the Navy. Therefore, it was selected for conversion to a destroyer 

seaplane tender to fill an identified capability gap until enough purpose-built tenders became 

available. Ballard was recommissioned in ordinary on 25 June 1940 to begin the process (Bruhn 

2016:19). Conversion work focused on removing offensive weaponry and two of the original 

four boilers to create space for tender-specific equipment. This expanded available deck space 

and added a 30,000-gallon (gal.) (113,562 liter [L]) aviation fuel tank, with lubricating oil 

storage, to allow fueling and servicing of seaplanes. It increased storage space to maintain 



   

 

33 
 

aviation-specific equipment, spare parts, munitions, and other necessary equipment. Two 30 ft. (9 

m) motor launches, each carrying 600-gal. (2,271 L) bowser tanks, were added to supplement 

seaplane refueling operations. Enhanced living spaces and messing facilities boosted life support 

for embarked aircrew.  

Ballard was equipped with new 500-lb. (227 kg) anchors as more modern and capable 

seaplane mooring anchors became available. Unfortunately, its anchor stowage racks were not 

designed for those anchors. The new heavier models did not fit below as intended and were 

instead stored on the open deck, posing safety and maintenance challenges. Ballard petitioned to 

return to smaller models and was reissued five 150-lb. (68 kg) Danforth and five Northill 200-lb. 

(91 kg) anchors, all then in stock at the shipyard (FIGURE 2.8). The small anchors, sufficient for 

smaller Curtiss-Wright SOC Seagull, Vought OS2U Kingfisher, and even PBY Catalinas in 

FIGURE 2.7: Photo of USS Ballard (DD-267) in 1920, two decades before its conversion to a seaplane tender 

(Photo courtesy U.S. Navy Historical Center #NH78139). 



   

 

34 
 

moderate weather conditions, were not intended to be used with an airplane as large as the 

Martin PBM Mariner.  

Ballard was then a seaplane tender in every sense of the word and was redesignated 

destroyer, seaplane tender (AVD) 10, or AVD-10 once this conversion was complete (FIGURE 

2.9). It was placed in full commission on 2 January 1941 and reported for duty to Aircraft 

Scouting Force on 5 March 1941. It was assigned to Fifth Fleet’s Northern Attack Force to 

FIGURE 2.8: Northill print advertisement for its compact, folding seaplane anchor (Image courtesy 9 May 1940 

Flight magazine). 



   

 

35 
 

support naval gunfire spotting floatplanes during Operation FORAGER and present at Saipan 

during the initial invasion on 15 June 1944. 

2.3.b. USS Onslow (AVP-48) 

USS Onslow was a purpose-built vessel constructed after the Navy recognized the need 

for additional seaplane tenders to support a burgeoning naval aviation patrol arm. Although 7 

Barnegat-Class small seaplane tenders (AVP) were authorized in 1938, the number grew to 31 by 

war's end. This class of vessel was named for small bodies of water and, in Onslow’s case, a 

concave indentation of the North Carolina Coast anchored by Cape Lookout in the north and 

FIGURE 2.9: USS Ballard (AVD-10) at Mare Island Navy Yard 30 December 1943 as it looked before Operation 

FORAGER (Photo courtesy San Francisco National Archives, Mare Island Navy Yard Ship Files). 

https://en.wikipedia.org/wiki/Cape_Lookout_(North_Carolina)


   

 

36 
 

Cape Fear to the south. These bespoke tenders possessed capabilities that allowed them to 

perform many functions, chief among them tending seaplanes. They also performed at-sea 

aircraft salvage, escort duties for large tenders, and other duties such search and rescue.  

 Onslow incorporated the latest technology and designs to include aircraft and weapons 

repair shops; aircraft crew and support crew facilities; weapons and fuel storage capacities; 

operating range and ability to conduct operations independent of other fleet vessels; air defense 

and fire support capabilities; and the ability to navigate and turn around in shallow or restricted 

waterways (Bruhn 2016:21). The 24th vessel in the class, its hull was laid on 18 May 1942 and it 

was placed in full commission on 22 Dec 1943 for service in the Pacific (FIGURE 2.10). It 

departed Hawaii on 17 March 1944 and arrived at Majuro in the Marshall Islands on 4 April 

1944. Onslow joined Fifth Fleet in Eniwetok on 9 June 1944 and tended to VP-16’s PBM-3D 

Mariner aircraft there in anticipation of its imminent deployment to Saipan (FIGURE 2.11).  

FIGURE 2.10: Launching of the USS Onslow (AVP-48) at the Lake Washington Shipyards, Houghton, Washington 

(Photo courtesy of National Archives #80-G-216474). 

 

https://en.wikipedia.org/wiki/Cape_Fear_(headland)


   

 

37 
 

2.3.c. USS Pokomoke (AV-9) 

USS Pocomoke, the largest and most capable of the three tenders employed in Saipan’s 

open ocean seadrome, was a large seaplane tender (AV) intended to support medium (PBM 

Mariner) and heavy (PB2Y Coronado) seaplanes. It began as a Type C3 Liberty Ship, SS 

Exchequer, built by Ingalls in its Pascagoula, Mississippi yard (FIGURE 2.12). Exchequer, 

notable as the first ship to be entirely made with steel plates welded end to end and traditional 

overlapping plates riveted together, was delivered by Ingalls Corporation to the U.S. Maritime 

FIGURE 2.11: Side view of USS Onslow (AVP-48) at Saipan Island in the Marianas as seen from USS Pokomoke 

(AV-9) on 22 June 1944 (Photo courtesy of National Archives #80-G-238469). 

 



   

 

38 
 

Commission on 16 October 1940 and then by the commission to the U.S. Navy on the same day 

at Mobile, Alabama, to begin its conversion (Anderson 2012).  

Pokomoke’s capabilities were significantly greater than those of Onslow and Ballard. It 

lacked the versatility of the smaller tenders, but it provided substantially more capability as a 

large seaplane tender. Its deck could facilitate three Seagull or Kingfisher Observation Seaplane, 

United Aircraft Division (OSU2) floatplanes or one larger patrol seaplane, including the PBM 

FIGURE 2.12: SS Exchequer in Pascagoula, Mississippi 1940 (Photo courtesy of Ingalls Corporation).  



   

 

39 
 

Mariner, for maintenance or salvage (FIGURE 2.13). It possessed substantial aviation fuel tanks, 

spare parts capacity, work and machine shops, and berthing space sufficient for an entire patrol 

squadron. In short, Pokomoke provided a singular seaplane tender capability. 

 

It was completed, fully mission-capable, and combat-ready on 30 November 1941, only a 

week before the U.S. was thrust into war. Pokomoke was initially assigned to the Atlantic Fleet 

but transferred to the Pacific and Fifth Fleet as the importance and demands of the Pacific War 

grew. It was forward-based to Eniwetok for Operation FORAGER in anticipation of its 

deployment to Saipan, where it played a pivotal role in supporting VP-16 operations in the open 

ocean seadrome (FIGURE 2.14).  

FIGURE 2.13: Diagram of large seaplane tender deck configurations. Top right depicts Pokomoke; the single crane 

and deck space for just one large seaplane is evident. On the left, the Curtis and Carrituck classes sport multiple 

cranes and deck space for two large seaplanes (Image courtesy Michael Kammen).  



   

 

40 
 

2.4. SAIPAN’S OPEN OCEAN SEADROME  

  

 Establishing the open ocean seadrome was not a preplanned activity but an expedient 

response to an emergent operational challenge. Although some planning had gone into 

developing an on-order seadrome later in the campaign in Saipan’s sheltered waters, primarily 

for refueling spotter float planes, these initial steps played heavily into the ability to respond to 

this more significant task in short order. Notably, a destroyer seaplane tender was assigned to the 

Northern Invasion Force and was immediately available for service. Previously conducted 

intelligence preparation allowed U.S. Navy planners to adjust to changing circumstances rapidly.   

 

 FIGURE 2.14: USS Pokomoke (AV-9) at Kwajalein, Marshall Islands. This photograph was taken on 9 June 1944, 

just eight days before Pokomoke would arrive at Saipan (Photo courtesy of National Archives #80-G-265980). 



   

 

41 
 

2.4.a Planning 

 

There was no plan, nor even passing consideration, for establishing a seadrome in the 

open ocean when the Marines stormed ashore onto Saipan on 15 June 1944. Fifth Fleet carrier-

based aircraft were relied upon exclusively for airborne ASW activities and surface searches. 

The large PBMs planned to forward deploy only after Tanapag Harbor and its sheltered lagoon 

were secured and an advance base established later in the battle (FIGURE 2.15). This base was 

anticipated to meet the conditions to provide a permanent facility offering the full range of 

seaplane service requirements for employing large amphibious patrol aircraft consistent with 

advance base doctrine. Until then, conventional wisdom held that there was no way to deploy or 

support large seaplanes in Saipan. Six seaplane tenders, including Pocomoke and Onslow, and 

several patrol squadrons, including VP-16, were forward deployed to Eniwetok and Kwajalein 

Atoll. Those ships anticipated deploying on order to Saipan once conditions for establishing the 

seadrome were met, followed by the patrol squadrons once facilities were able to receive them. 

Navy planners recognized the operational necessity of a temporary seaplane base to 

service smaller OSU2 Kingfishers and SOC Seagull floatplanes assigned to battleships and 

cruisers as naval gunfire spotters. By untethering floatplanes from their parent ships and 

assigning support functions to a designated seaplane area, the bombardment ships could stay on 

the gunline, focused on Japanese targets instead. VADM Turner, Commander, Task Force 51, 

prepared a detailed plan as Annex (M) of Operation Plan A10-44 for both a “Seaplane Tender 

Base” or an “Improvised Seaplane Base,” depending upon resources available (COM 5th 

PHIBFOR War Diary 1944:309). This was to be executed “when the situation permitted the 

planting of mooring buoys in sheltered water, the anchoring of the seaplane base vessel close by, 

and the safe operation of seaplanes from the area” (COM 5th PHIBFOR War Diary 1944:310).  



   

 

42 
 

Ballard was assigned to the Northern Assault Force as the “seaplane base vessel” to 

fulfill this mission. This had the fortunate effect of placing it off the invasion beaches on the first 

day of the Battle of Saipan instead of 1,000 mi. (1,609 km) away in Eniwetok. Although the 

temporary seaplane base concept never came to fruition as events quickly overcame plans, 

Ballard was in place and ready to accept alternate tasks. It did not have to wait long. 

 

2.4.b. Establishment and Operations of the Open Ocean Seadrome 

The unplanned participation of U.S. Navy Martin PBM-3D Mariners in the early days of 

Operation FORAGER was a direct response to the unanticipated reaction of the Japanese First 

FIGURE 2.15:  The bombing of the Japanese seaplane base at Flores Point, Saipan, by plane from USS White Plains 

(CVE-66), July 3, 1944. Seaplane ramps visible in the center of this photo would be in use by PBM-3D Mariners 

just seven days later (Photo courtesy of National Archives #80-G-384164). 



   

 

43 
 

Mobile Fleet. Intelligence estimate continued to suggest, “Major units of the Japanese Fleet are 

capable of opposing our operations against FORAGER. It is believed that these units of the 

Japanese Fleet will not interfere with our operation against FORAGER except for “hit and run” 

raids on detached units of our forces, but will conserve their strength until a time when they can 

go into action in waters nearer their main bases” (COM 5th PHIBFOR 1944:137). Academics 

and civilian analysts reinforced that the IJN would not seek a fleet engagement to senior 

leadership (Kirafly 1943:45).  

VADM Spruance was caught off guard when the Japanese First Mobile Fleet contested 

the U.S. invasion. In retrospect, it is difficult to imagine that the intended conquest of the 

Marianas that brought the Japanese home islands within range of America’s newest long-range 

heavy bomber, the B-29 Superfortress, could not provoke such a reaction. The Commander, Fifth 

Fleet, responded by ordering six VP-16 PBM-3Ds forward. These aircraft were to conduct night 

surface RADAR searches up to 700 mi. (1,127 km) west of Saipan, far beyond the more limited 

range of carrier aircraft, and provide the Fifth Fleet with vital intelligence on the First Mobile 

Fleet. Two additional tenders, Pokomoke and Onslow, were also ordered to Saipan to support 

them. 

Meanwhile, Ballard, already at Saipan, was immediately pressed into service to establish 

the open ocean seadrome. It was to detach from its task group and proceed to the Saipan Shelf on 

the evening of 16 June 1944 to set moorings and buoys for the Mariners’ arrival the following 

morning. Ballard worked through the night and was ready to receive the PBMs by the morning 

of 17 June 1944. It may be assumed that the anchors used were the reissued Northill 200-lb. (91 

kg) and Danforth 150-lb. (68 kg) anchors, as there is no record of Ballard altering its anchor 

storage lockers to accommodate larger anchors.  



   

 

44 
 

The selected offshore location sought to balance suitability for seaplane operations with 

protection from Japanese attack (USS Ballard History 1945:10). The area, 4 mi. (6.5 km) 

offshore in water 100 ft. (31 m) deep, performed neither function well. Waves were excessively 

high that far out with no shelter to break the high winds, yet it was still within the far reaches of 

Japanese shore-based artillery and coastal defense guns. There was no elegant solution available 

to solve this wicked problem. However, the selected location provided reasonable protection 

from the Japanese while still allowing the PBMs to operate, albeit outside their normal limits. 

These suboptimal conditions were accepted, given the importance of the long-range RADAR 

search mission to the security of the Fifth Fleet. It was the first time the Navy had attempted to 

operate from such an innovative open ocean seadrome.  

Aircraft and aircrew relied upon seaplane tenders for all their needs, given the complete 

absence of terrestrial support facilities; the importance of the tenders to seaplane operations 

cannot be overstated. Billeting and messing spaces for aircrew, maintenance teams, replacement 

parts for aircraft, transportation to and from moored aircraft, and availability of fuel and oil to 

feed the engines during their 14-hour-long patrols were all wholly dependent upon the tenders. 

Ballard was initially responsible for all these tasks and more.  

Although Ballard planned to support the small OSU2 Kingfisher of SOC Seagull 

floatplanes, it became responsible for supporting the inaugural detachment of six PBM-3D 

Mariners scheduled to arrive on 17 June 1944. The Mariner was a beast compared to small 

floatplanes. Ballard possessed just 30,000 gal. (113,562 L) of aviation fuel while the PBM tank 

capacity was 2,914 gal. (11,030 L); as such, it could only refuel the six initially deployed PBM 

aircraft perhaps twice before depleting its stores. It had no PBM spare parts or personnel required 

to conduct significant maintenance on the Mariners. It could not have reloaded bombs, 



   

 

45 
 

torpedoes, or depth charge munitions even if they were available. Yet this lone, insufficiently 

capable tender did, by necessity, become the foundation for the open ocean seadrome. Few 

additional berthing spaces were available aboard Ballard, and the 60 crewmen that comprised 

this Mariner complement could not be accommodated appropriately, forcing many men to live 

aboard the aircraft. 

The maritime environment from which the aircraft were to operate was equally crucial to 

seaplane operations as the tender support received. An expeditionary forward seaplane base was 

usually established in a sheltered harbor or lagoon to protect the planes, and tenders from the 

worst nature had to offer. These protected waters should optimally have been deep enough not to 

interfere with seaplane taxiing or takeoff but sufficiently shallow to facilitate the easy placement 

of seaplane moorings and runway markings, while mitigating the size of the waves and ocean 

swell. Although ruggedly built, patrol seaplanes were still vulnerable to excessive sea conditions 

and could be damaged or worn down by the continuous beating of a heavy sea. In sum, “Weather 

conditions had a far more serious impact upon flying boat operation than they had on operation 

of land-based aircraft assigned to fixed airstrips” (Kammen 1958:68). 

Seaplane flight operations necessitate limits to wind and waves to take off and land 

safely. Excessive sea state reduced fuel and weapons load an aircraft could sortie with to lighten 

the takeoff load, thus reducing range and offensive potential. They also risked damage to the 

aircraft and injury to the crew. An appropriate seadrome needed sufficient ramps and 

maintenance areas so a seaplane could be fully pulled out for inspection and maintenance. A 

ramp was vital if an aircraft was damaged in battle or a routine accident. Even the most minor 

damage to an aircraft could result in catastrophic loss if it began taking on water and could not be 

safely hoisted or beached; this reality was dramatically reinforced at Saipan. 



   

 

46 
 

Saipan’s open ocean seadrome had none of these features. Instead, it was established well 

offshore beyond the outer reef, initially 4 mi. (6.5 km) seaward. The seaplanes needed to operate 

this far offshore to remain out of small arms and light-caliber artillery range as the Japanese 

occupied the whole island when they deployed. Regardless, they remained under the guns of 

heavier Japanese coastal defenses and artillery until the Marines pushed the Japanese sufficiently 

far inland to neutralize this threat. Takeoffs from the leeward side of the island into the wind 

took them closer to the Japanese stronghold, making them vulnerable to anti-aircraft fire. 

Japanese air attacks also targeted the PBMs, beginning just hours after their arrival (VP-16 

History 1945:37). 

Enemy actions were only one of the many challenges, the water depth being another. The 

Saipan Shelf's water depth sometimes exceeded 100 ft. (31 m), pushing mooring capabilities to 

their limits and precluded using marker buoys to designate takeoff and landing areas. This was 

especially critical at night when many operations were conducted because lighted buoys could 

not be permanently emplaced to guide the airplanes, and illuminating ships to assist remained 

dangerous under the Japanese guns.  

The wind and sea state determined takeoff direction and taxiing requirements. 

Established in the northernmost portion of the transport anchoring area and maximally away 

from surface vessel traffic, there was never enough room to maneuver. The risk of striking 

another ship or seaplane, especially under the blackout conditions normally operated under, was 

always high. A collision with a surface vessel disabled a Mariner on the first night. Errant ships 

and boats, floating debris, and numerous derelict and wayward landing craft compounded the 

hazards to these aircraft (USS Alhena 1944:7). 



   

 

47 
 

The impact of the weather and sea state was felt immediately upon the arrival of the 

Mariners in Saipan on the afternoon of 17 June 1944. They found “man-sized” waves awaiting 

the