USF Libraries
USF Digital Collections

Assessing the ability of soils and sediment to adsorb and retain cs-137 in puerto rico


Material Information

Assessing the ability of soils and sediment to adsorb and retain cs-137 in puerto rico
Physical Description:
Ithier-Guzman, Warner
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:


Subjects / Keywords:
Dissertations, Academic -- Marine Science -- Masters -- USF   ( lcsh )
non-fiction   ( marcgt )


ABSTRACT: As part of the radioactive exercises taking place around the world anthropogenic radionuclides were introduce to Puerto Rico's terrestrial and aquatic environments beginning in 1962. Two major projects took place in the island, the Rain Forest Project and the construction of a Boiling Superheat Nuclear Power Plant (BONUS). While in operation several accidental shutdowns occurred at the BONUS facility. One of these accidental shutdowns released 582 MBq into the nearby environment. Vieques an island located few miles east of the main island has received anthropogenic inputs of heavy metals resulting from military practices conducted by the US Navy. Due to the potential presence of Cs-137 in soils and sediments in Puerto Rico a radiological assessment was performed. Downcore soil and sediment analysis as well as surface samples analysis was conduct in these three sites indicating the presence of Cs-137. Activity range varies among site from below detection limit to 12 dpm/g at Vieques, 15 dpm/g at Espiritu Santo Estuary and 12 dpm/g at the BONUS Facility. ICP-OES analysis indicates the existence of an oxic environment at the sedimentary system of the island. Cs-137 retention is strongly influenced by particle grain size and at the study sites clay was present in less than 20% for most sites. An X-ray diffraction analysis show that kaolinite and smectite are present at all sampling sites and illite is absent. To further analysis the ability of soil and sediments to retain adsorption and desorption was conducted using clay reference material and samples from the island. All samples, reference and natural, used in the study were placed in an aqueous solution that contained MES buffer (5.0 micromol, pKa of 6.1), ammonium nitrate (0.010 M) and the five metals (individual concentrations ranged from 0.48 micromol to 1.6 micromol). Solution pH was adjusted by titration with acid or base, depending on the nature of the sample. Results were quantified as distribution coefficients. These results indicate that the absorption and retention of Cs-137 in the sediments in Puerto Rico is driven by the mineralogy of the site.
Dissertation (PHD)--University of South Florida, 2010.
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Warner Ithier-Guzman.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains X pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - E14-SFE0004637
usfldc handle - e14.4637
System ID:

This item is only available as the following downloads:

Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2010 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0004637
XX9999 (Online)
1 100
Ithier-Guzman, Warner.
0 245
Assessing the ability of soils and sediment to adsorb and retain cs-137 in puerto rico
h [electronic resource] /
by Warner Ithier-Guzman.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains X pages.
Dissertation (PHD)--University of South Florida, 2010.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
3 520
ABSTRACT: As part of the radioactive exercises taking place around the world anthropogenic radionuclides were introduce to Puerto Rico's terrestrial and aquatic environments beginning in 1962. Two major projects took place in the island, the Rain Forest Project and the construction of a Boiling Superheat Nuclear Power Plant (BONUS). While in operation several accidental shutdowns occurred at the BONUS facility. One of these accidental shutdowns released 582 MBq into the nearby environment. Vieques an island located few miles east of the main island has received anthropogenic inputs of heavy metals resulting from military practices conducted by the US Navy. Due to the potential presence of Cs-137 in soils and sediments in Puerto Rico a radiological assessment was performed. Downcore soil and sediment analysis as well as surface samples analysis was conduct in these three sites indicating the presence of Cs-137. Activity range varies among site from below detection limit to 12 dpm/g at Vieques, 15 dpm/g at Espiritu Santo Estuary and 12 dpm/g at the BONUS Facility. ICP-OES analysis indicates the existence of an oxic environment at the sedimentary system of the island. Cs-137 retention is strongly influenced by particle grain size and at the study sites clay was present in less than 20% for most sites. An X-ray diffraction analysis show that kaolinite and smectite are present at all sampling sites and illite is absent. To further analysis the ability of soil and sediments to retain adsorption and desorption was conducted using clay reference material and samples from the island. All samples, reference and natural, used in the study were placed in an aqueous solution that contained MES buffer (5.0 micromol, pKa of 6.1), ammonium nitrate (0.010 M) and the five metals (individual concentrations ranged from 0.48 micromol to 1.6 micromol). Solution pH was adjusted by titration with acid or base, depending on the nature of the sample. Results were quantified as distribution coefficients. These results indicate that the absorption and retention of Cs-137 in the sediments in Puerto Rico is driven by the mineralogy of the site.
Advisor: Ashanti Johnson, Ph.D.
Dissertations, Academic
x Marine Science
t USF Electronic Theses and Dissertations.
4 856


Assessing the Ability of Soils and Sediment to Adsorb and Retain Cs-137 in Puerto Rico by Warner Ithier-Guzman A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy College of Marine Science University of South Florida Major Professor:Ashanti Johnson, Ph.D. Robert Byrne, Ph. D. Kathleen Carvalho, Ph.D. Brandon Jones, Ph.D. Edward VanVleet, Ph.D. Date of Approval: July 14, 2010 Keywords: radionuclide, monitoring, so il, mineralogy, adsorption, desorption Copyright 2010 Warner Ithier-Guzmn


DEDICATION Pursuing a doctoral degree was the biggest dream I ever had in my life. As the first member of my family to go to college, th is challenge was more than a risk, it was an adventure. This adventure was made possi ble by the opportunity I received from Dr. Ashanti Johnson. After knowing each other for a few years, she did not hesitate to offer me a position in her laboratory seven years a go. That opportunity I was given has made the difference in my life and in my family. I knew it would not be easy and that I would have to make many sacrifices to reach my goal, but I accepted the challenge and on August 17, 2003 the journey to the Ph.D. began. Today I am grateful that Dr. Johnson introduced me to the fascina ting world of geochemistry. This new world has provided me with knowledge and skills need ed to achieve my dreams. Along my journey I found great people that supported my efforts in achieving my goal of obtaining a doctoral degree. These pe ople have made a difference in my life. Some of these supporters were members of my committee and I want to thank them for their continuous support and gui dance. I was honored to have Dr. Robert Byrne, Dr. Kathleen Carvalho, Dr. Edward VanVleet a nd Dr. Brandon Jones, along with Dr. Ashanti Johnson serve on my committee. At several stages along my journey these individuals were extremely intimidating, particularly dur ing my proposal defense, my comprehensive exams and the final process of completing and ultimately defending my dissertation. These same individuals have also provided me with moments of happiness, moments that


I will remember forever. One such mo ment happened on September 6, 2007 when Dr. Byrne congratulated me on passing my compre hensive exams. He knew that it was a major relief for me to hear him say that I pass ed, since I had been ab solutely terrified of his CO2-related questions. During my time at the University of Sout h Florida College of Marine Science, I learned not only about CO2 systems, organic matter degrad ation, radiogeochemistry, etc., I also experienced life. Saint Petersburg, Florida became my home and new friends came into to my life. It was some of these frie nds that provided me unc onditional support that helped me survive this process. I called on these friends when I felt as though I had no hope of ever finishing. I can not adequately express the gratitude I have for these friends and my extended "USF family" which include Guillem Mateu, Marta Rodriguez, Karyna Rosario, Michael Martinez and Camille Daniel s; my lab mates, Nekesha Williams and Patrick Schwing, as well as se veral other individuals that started with me and decided to go another route on their life journey. Thes e individuals supported and encouraged me when I suffered, joined me in celebrating my accomplishments and laughed with me during special moments of joy. In short they we re there for me when I needed them. For this I am forever grateful. Their friendship is one of the most importa nt things that I will take along wherever the next phase of my j ourney leads me. I also want to take a moment to thank a person I consider to be more than a friend, he is like a brother, Esteban Martinez. Esteban and his family we lcomed me into their home and provided me with great Cuban food every nigh t while I worked on the final stages of my dissertation. There are many other friends that I met th rough the Florida-Georgia LSAMP Bridge to the Doctorate Program and the MSPHDS Progr am which made my stay at USF easier


and happier. To those members of my "Bridge to the Doctorate family and the MSPHDS Family" I offer my heartfelt thanks. While at the University of South Florid a I also had the opportunity to educate others and serve as a mentor for many unde rgraduate interns who provided assistance with my research. Those interns were; Jay ce G, Carlos Rosa, Maruiz Marrero, Carmen Berrios, Adriana Quijano and Kemmy Oguns anwo. All of thes e interns made a difference by contributing intellectually a nd became examples for others in the geosciences field. I am also grateful for the technical a nd intellectual support I received from Dr. Kelly Quinn, Dr. Donny Smoak, Dr. Crawford Elliot and Dr. Claudia Benitez-Nelson. The knowledge they shared was truly meaningful I also would like to recognize the help I received from the USGS for facilitating my training and operation of the X-Ray diffraction instrument and to Dr. Byrne for f acilitating the use of one of the instruments needed to conduct this research. This project is the result of a collect ive effort from many people and other institutions on the island of Puerto Rico; therefore I want to thank the Vieques National Wildlife Refuge Staff and Lymari Orellana, Migdalia Ruiz and Jorge Ramos from the University of Puerto Rico, Rio Piedras. Last but not least, I want to thank my family for the support I received from them. My mother, who served as my inspirat ion, and also my dear sister Vanessa, who was always there to listen to me during hard moments. It was my mom’s determination, dedication and encouragement that influen ced my decision to pursue this degree and today I want to reiterate how mu ch I love her and thank her for giving me life. I am also


grateful to, Jaime; his emoti onal support was critical during my time at the University of South Florida. Jaime always had words of wisdom for me when I experienced difficult moments, and when I was about to give up. Without my family I would not have been able to reach this dream.


i TABLE OF CONTENTS LIST OF FIGRES ........................................................................................................ iii LIST OF TABLES ....................................................................................................... vi ABSTRACT ................................................................................................................ vii CHAPTER I ENVIRONMENTAL MONITO RING OF CS-137 IN THE CARIBBEAN AND HYDROGEOLOGY OF PUERTO RICO INTRODUCTION .............................................................................................2 Geology and soil distribution of Puerto Rico .......................................4 Hydrogeology of study sites ...............................................................11 Studies on Cs-137 mobility .................................................................12 Sources of radiation in the environment .............................................14 Nuclear weapon testing .......................................................................15 Environmental risk caused by anthropogenic radionuclides in the Caribbean Region......................................................................16 Health risk associated with radiation ..................................................18 STUDY AREA ................................................................................................20 Boiling Nuclear Superheated Reactor ..................................................20 Atlantic Fleet Weapons Training Facility ............................................21 Caribbean National Forest ...................................................................23 HYPOTHESES ..............................................................................................25 OBJECTIVES ..................................................................................................26 CHAPTER II RADIOGEOCHEMISTRY OF THREE STUDY SITES IN PUERTO RICO INTRODUCTION ...........................................................................................28 Vieques ................................................................................................28 Boiling Nuclear Superheated Power Station…………………………29 El Verde Experimental Station …..……………………….………….32 The Rainforest Project………………………………………………..34 METHODS ......................................................................................................39 Gamma analysis for Cs-137 .................................................................43 Gamma analysis of Pb-210 and Ra-226...............................................44 Grain size analysis ...............................................................................44 Loss on ignition analysis ......................................................................45 Inductively Coupled Plasma-O ptical Emission Spectrometer .............46 Mineralogical analys is X-ray diffraction .............................................47


ii RESULTS ........................................................................................................49 Gamma spectroscopy ...........................................................................49 Grain size analysis ...............................................................................61 Lost on ignition (LOI) .........................................................................68 Inductively Coupled Plasma-O ptical Emission Spectroscopy (ICP-OES) ...........................................................................................72 X-Ray Diffraction ................................................................................78 DISCUSSION ..................................................................................................81 CONCLUSIONS..............................................................................................90 CHAPTER III ADSORPTION AND DESO RPTION OF RUBI DIUM, COPPER, CADMIUM, CESIUM AND LEAD ON CLAY REFERENCE MATERIAL INTRODUCTION ...........................................................................................96 METHODS ......................................................................................................99 Experimental Methods .......................................................................102 Desorption Experiments.....................................................................103 Sorption Kinetics ...............................................................................104 Sorption Experiments.........................................................................104 Analysis..............................................................................................105 RESULTS ......................................................................................................107 Desorption experiments (clay minerals) ............................................107 Equilibration kinetics (clay minerals) ................................................113 Sorption on reference material ...........................................................115 Sorption results for soils and sediments in Puerto Rico ....................118 DISCUSSION ................................................................................................122 CHAPTER IV GEOCHEMICAL FACTORS AFFECTING RESULTS GEOCHEMISTRY OF THE STUDY SITES AND HOW THEY IMPACT CURRENT RESULTS...................................................................................126 MAJOR CONCLUSIONS .........................................................................................130 REFERENCES ..........................................................................................................135 APPENDICES ...........................................................................................................146 Appendix A. Gamma spectroscopy calibration ............................................147 Appendix B. X-ray diffraction calibration curve ..........................................148 Appendix C. Loss on ignition protocol .........................................................149 Appendix D. EPA Method 3050b .................................................................150 Appendix E. Sample Preparation for XRD Analysis ....................................153 Appendix F. Data from adso rption/desorption experiments .........................155 Appendix G Self-adsorption and weight corrections .....................................162 ABOUT THE AUTHOR ............................................................................ END PAGE


iii LIST OF FIGURES Figure 1 Geological formations of Puerto Rico…... …………………………..…...5 Figure 2 Assumptions for the extrapolati on of the cancer risk vs. radiation dose to low-dose levels, given a known risk at a high dose..………..…..18 Figure 3 BONUS Reactor sampling area..……………………………….………..21 Figure 4 Vieques sampling site…………………………………….……….….….22 Figure 5 Sampling sites at El Verde a nd the Espiritu Santo Estuary. Rio Grande, Puerto Rico..……………………………………………….…....24 Figure 6 Sampling sites in Puerto Rico……………………………………………40 Figure 7 Surface and downcore Samp ling Sites in Vieques, PR..……….………..41 Figure 8 Downcore Cs-137 and ex Pb-210 profiles at Kiani Lagoon, Vieques, PR…………...............................................................................50 Figure 9 Downcore Cs-137 and ex Pb-210 profiles at Mosquito Bay, Vieques, PR……………………………………...............................51 Figure 10 Gamma analysis and clay c ontent of surface samples, Vieques…………51 Figure 11 Downcore Cs-137 and exPb-210 profiles at BONUS St. 1…………...…52 Figure 12 Downcore Cs-137a nd exPb-210 at BONUS St.2, Rincn, PR………….53 Figure 13 Downcore Cs-137and exPb -210 at BONUS St. 3, Rincn, PR………….54 Figure 14 Downcore Cs-137and exPb -210 at BONUS St. 4, Rincn, PR………….54 Figure 15 Downcore Cs-137 and exPb-210 at El Verde St. 1 Rio Grande, PR..…...55 Figure 16 Downcore Cs-137and exPb-210 at El Verde St. 2 Rio Grande, PR……..56 Figure 17 Downcore Cs-137 and exPb-210 at Espiritu Santo Estuary St.1 Rio Grande, P.R………………………………………………………………57 Figure 18 Downcore Cs-137 and exPb-210 at Espiritu Santo Estuary St.2 Rio Grande, P.R………………………………………………………………58


iv Figure 19 Cs-137 Inventories for Stations at the El Verde and Espiritu Santo Estuary…………...………………………………………………………60 Figure 20 Correlation between Cs-137 activities on surface samples from the radiat ion center at El Verde Experimental Station and fine r particle size distribution .………………60 Figure 21 Grain size dist ribution at Kiani Lagoon Station 1, Vieques, PR…...……61 Figure 22 Grain size distributions at Mosquito Bay Lagoon Station 1, Vieques, PR…….………………………………………...………………62 Figure 23 Grain size distribution at BONUS Area, Station 1, Rincn, PR…………63 Figure 24 Grain size distribution at BONUS Area, Station 2, Rincn, PR…………63 Figure 25 Grain size distribution at BONUS Area, Station 3, Rincn, PR…………64 Figure 26 Grain size distributions at BONUS Area, Station 4, Rincn, PR………..64 Figure 27 Grain size dist ribution at Espiritu Sa nto Estuary St. 1, Rio Grande, PR……………………………………………………………….65 Figure 28 Grain size dist ribution at Espiritu Sa nto Estuary St. 2, Rio Grande, PR……………………………………………………………….66 Figure 29 Grain size distribution at El Verde St. 1, Rio Grande, PR……………….66 Figure 30 Grain size distribution at El Verde St. 2, Rio Grande, PR……………….69 Figure 31 organic matter percenta ges at Kiani Lagoon and Mosquito Bay, Vieques, PR………………………...………………………………70 Figure 32 Organic matter percenta ge at BONUS Area, Rincn, PR……………….71 Figure 33 Organic matter in percentage at El Verde and Espiritu Santo Estuary, Rio Grande, PR.……………………...…………..………………………71 Figure 34 Organic matter in percenta ge at El Verde Surface Samples……………..71 Figure 35 Downcore concentratio n for Iron and Manganese at Mosquito Bay St. 1, Vieques………………………....……………….....73 Figure 36 Downcore concentratio n for Iron and Manganese at Kiani Lagoon St. 1, Vieques.…….………….………….………………..73


v Figure 37 Iron and Manganese concentr ation for surface samples at Vieques National Fish and Wildlife Refuge, Vieques, PR ….……………………73 Figure 38 Downcore concentration for Iron and Manganese at BONUS Area St. 3, Rincn, PR. ………….…………..………….………………74 Figure 39 Downcore concentration for Iron and Manganese at BONUS Area St. 4, Rincn, PR. ………………………………………....……………...75 Figure 40 Downcore concentratio n for Iron and Manganese at El Verde St. 1, Rio Grande, PR…….……………………….…..……….75 Figure 41 Downcore concentratio n for Iron and Manganese at Espiritu Santo Estuary St. 1, Rio Grande, PR……………….…………...76 Figure 42 Downcore concentratio n for Iron and Manganese at Espiritu Santo Estuary St. 2, Rio Grande, PR…………….……..….…....77 Figure 43 Mineralogical analysis of bulk samples from Vieques, PR……..……….78 Figure 44 Mineralogical analysis of bulk samples from the BONUS area, Rincon, PR ……………………………………………….79 Figure 45 Mineralogical analys is of bulk samples from El Verde and Espiritu Santo Estuary………………….………………....80 Figure 46 Cs-137 fallout record for the Miami region, adapted from Surface Air Sampling Program, Department of Homeland Security, 2008…….………….………….……………………86 Figure 47 Sampling locations in Puerto Rico ……….………….…………..……..101 Figure 48 Desorption experime nt on clay minerals ……….….…….………….....109 Figure 49 Desorption experiments on soils and sediments from Puerto Rico..........................................................................................................111 Figure 50 Element concentration over time on clay minerals……………………..114 Figure 51 Metal adsorption on clay mineral…………………...………………….116 Figure 52 Metal behavior in soils and sediments from different study sites ……..120


vi LIST OF TABLES Table 1 Soil type and soil series at study sites…………………………………..7 Table 2 Laptev Sea-Lena River es tuary region average sediment Cs-137 activity (Bq kg-1) and grain size distribution.………………...13 Table 3 Past activities (measured in 1968) and recent activity estimates (determined in 2001) of principal radionuclides entombed with the BONUS reactor for 1968 and 2001 ......………...….31 Table 4 Past activities (measured in 1968) and recent activity estimates (determined in 2001) of principal radionuclides external to the entombed BONUS reactor for 1968 and 2001…………………………….31 Table 5 Summary of radionuclides used at the Rainforest Project…………...…36 Table 6 Radionuclide movements on ve getation after th eir injection (Activity in Bq/g dry weight)……………………..……………………37 Table 7 Sampling location, descri ption of habitat conditions …………….……42 Table 8 Cation exchange capacity of clay minerals ……………...…………...100 Table 9 Initial concentration of metals added to adsorption experiment……………………………………………………...……105 Table 10 Summary of results of desorption experiments on clay minerals …………………………………………………...……110 Table 11 Summary of results of desorption experiments for soils and sediments in Puerto Rico ………………………………………..112 Table 12 Summary of results of adsorption experiments on clay minerals………………………………………………….………117 Table 13 Experimental pH for each soil type ………………….........................119 Table 14 Summary of results of adsorption experiments for soils and sediments in Puerto Rico ………………………………….121


vii Assessing the Ability of Soils and Sediments to Adsorb and Retain Cs-137 in Puerto Rico Warner Ithier-Guzman ABSTRACT As part of the radioact ive exercises taking place around the world anthropogenic radionuclides were introduce to Puerto Rico ’s terrestrial and aquatic environments beginning in 1962. Two major proj ects took place in the islan d, the Rain Forest Project and the construction of a Boiling Superheat Nuclear Power Plant (BONUS). While in operation several accidental shutdowns occurre d at the BONUS facility. One of these accidental shutdowns released 582 MBq into the nearby environment. Vieques an island located few miles east of the main island ha s received anthropoge nic inputs of heavy metals resulting from military practices c onducted by the US Navy. Due to the potential presence of Cs-137 in soils and sediments in Puerto Rico a radiol ogical assessment was performed. Downcore soil and sediment analysis as well as surface samples analysis was conduct in these three sites indicating the presence of Cs -137. Activity range varies among site from below detection limit to 12 dpm/g at Vieques, 15 dpm/g at Espiritu Santo Estuary and 12 dpm/g at the BONUS F acility. ICP-OES analysis indicates the existence of an oxic environment at the sedimentary system of the island. Cs-137 retention is strongly influenced by particle grain size and at th e study sites clay was


viii present in less than 20% for most sites. An X-ray diffraction analysis show that kaolinite and smectite are present at all samp ling sites and illite is absent. To further analysis the ability of soil and sediments to retain adsorption and desorption was conducted using cl ay reference material and samples from the island. All samples, reference and natural, used in the study were placed in an aqueous solution that contained MES buffer (5.0 micromol, pKa of 6.1), ammonium nitrate (0.010 M) and the five metals (individual concentrations ranged from 0.48 micromol to 1.6 micromol). Solution pH was adjusted by titration with acid or base, depending on the nature of the sample. Results were quantified as distribution coefficients. These results indicate that the absorption and retention of Cs-137 in the sediments in Puerto Ri co is driven by the mineralogy of the site.




2 INTRODUCTION Environmental monitoring of Cs-137 afte r the nuclear era and the Chernobyl accident in 1986 has been performed in many countries around the world. Monitoring of Cs-137 as well as other radionuclides is th e result of the intr oduction of manmade radionuclides into the environment primarily from nuclear powered weapons testing activities and nuclear po wer plant accidents (UNS CEAR 1982, Aakrog, 1994, Avery, 1996). These activities were c onducted by the United States and the former Soviet Union in a variety of environmental settings. Lo cations for testing incl uded the Hanford Site, WA (US); Savannah River, GA (US); Yucca, AZ (US); Novaya Zemlya (Russia) and Siberia (JohnsonPyrtle, 1999; Robinson and Noshkin, 1999; Myasoedov, 2000; WHO, 2001; Uyttenhove et al; 2002; Aarkrog, 2003). Th e testing of nuclear weapons was not restricted to the continents; several islands including the Marshall Is lands, Fangatufa and Bikini Atoll and the Caribbean is land of Puerto Rico were also utilized for this type of testing (PRNC, 1970, US Navy; 2002, Moon et al 2003). Monitoring of Cs-137 is useful to understand global fallout processes as we ll as for use as a chronological and dating tool. Many scientists use Cs-137 and other radionuclides (Pb-210) as chronology tools for sediment dating and as environmental tracers. Studying the activity of Cs-137 in places where anthropogenic input is known, allows us to calculate external inputs resu lting from human activity as well as, the activity resulting from fallout and its reten tion in depositional systems. Continued


3 assessment will facilitate remediation work and help avoid further radio-contamination at specific sites, thus safeguarding these ecosystems. Although above ground nuclear testing ende d in 1996 after the adoption of the Comprehensive Test Ban Treaty, residual e ffects may be important as a result of significant levels of radiation fallout in some areas of th e globe, especially in the Northern hemisphere (UNSCEAR 2001). Deforestation, dest ruction of habitat, water quality degradation, increased erosion and sedimentation, all have the potential to transport radionuclides from their original site and deposit them in areas where they may not only threaten natura l resources, but also pose significan t human health risks (Baqar et al. 2003; Hunter and Arbona, 1995). The current study focuses in the retention of Cs-137 by soils and sediments in the island of Puerto Rico. Performing this assessm ent in Puerto Rico will provide the only data available on the island for this type of research. Contributions to advance the science of radiogeochemistry on the island ar e very important, as its population growth and associated development projects (both re sidential and industrial) are impacting the island’s natural res ources. Therefore knowing whether pa st governmental activities had any type of impact on local eco system is essential. Municipalities such as Rio Grande, Rincn and Mayaguez were the main study sites for the Puerto Rico Nuclear Centre (PRNC) in the early 1960’s. The m unicipality of Rio Grande was also used by Odum to conduct his famous project on tropical irradiation at the Caribbean National Forest (PRNC, 1969). Odum’s project consisted of the irradiation of forest parcels to examine the effect of radiation on tropical ecosystems. Both projects, PRNC and the irradiation pr oject, were sponsored by the government, and


4 after the introduction of radi onuclides into the ecosystem, no further examination was conducted. The use of short lived isotopes as well as long lived isotopes is documented, but no further information on ecosystem status is known. This environmental monitoring will serve as the baseline information for further activities in Puerto Rico. Many studies have been conducted to study heavy metal concentration, polycyclic aromatic hydrocar bons (PAH), aerosols and Saharan dust, but no one has monitored the Cs-137 global fallout record or the retention of Cs-137 by sediments and soils in the place s where they were introduced. Geology and soil distribution of Puerto Rico The island of Puerto Rico, just as all is lands in the Greater and Lesser Antilles, has a complex geologic history. It has been es timated that Puerto Rico was formed more than 138 million years ago as a volcanic arc system (Larue, 1994). This arc formed due to subduction of the North American Plat e below the Caribbean Plate, which began during the Cretaceous and continued thr oughout the Eocene. Arc volcanism in the Greater Antilles ended after the collision of the Greater Antilles arc with the Bahamas Platform (Larue, 1994). The tectono-volcanic history of the island of Puerto Rico is found preserved in three main blocks. The Southwest, Centra l and Northeast litho-tectonic blocks have unique geochemical signals and distinct geologic histories (Larue, 1994). The geographic locations of the studied areas in reference to these bl ocks are significant because they have a control over pedogenesis and watershed development that will control soil formation and sediment transport.


5 The southwest block of the island has th e oldest rocks found in Puerto Rico. These are found in the Sierra Bermeja Complex and are composed mainly of serpentinites (Larue, 1994). Igneous rocks are also found in this complex having an oceanic island arc affinity, where amphibolite s have an oceanic crust signal (Larue, 1994). The rest of this block has a un ique array of depositional facies, both volcanoclastic and limestone in origin. Envi ronments range from basin to shallow water carbonate platforms (Larue, 1994). The oldest rocks in the Central block are found in the pre-Robles group composed of basin to shal low water carbonate platform environments similar to the previous block (Larue, 1994) A main volcanic center, Utuado Pluton, contains highly mineralized deposits (i.e., Cu) which, due to mining prospecting, creates areas that are impacted by acid mine drainage Vieques Island shares the same geologic history with this block, mainly related by volcanic rocks. The last block in Puerto Rico’s geology is the Northeastern block. Figure 1. Geological formation of Puerto Rico (Larue, 1994) The oldest rock comes from deepwater basinal deposit such as Daguao, Fajardo and Tabonuco among other formations. The Northeast block is characterized by abundant limestone formations. Unlike th e Central block, no pluton formations are


6 located in this region. The limestone formati on extends from west of the Rio Espiritu Santo to Aguadilla, the northwestern tip of th e island (Olcott, 1999). The length of this limestone formation is about 145 km and it r eaches its maximum width of about 23 km in the area of Arecibo, a town in the north central portion of Puerto Rico (Figure 1) (Olcott, P.G. 1999). The limestone in the island ha s experienced erosion by dissolution, and the limestone on the north is a good example of karst topography (Olcott, 1999). The historic volcanic activity resulted in an island that possesses mountain ranges that abruptly rise from the coast and valleys. For instance, the east-west trending Central Mountain Range divides the island into north and south and covers almost half of the island area with an average elev ation of 915 meters. The Sie rra de Luquillo rainforest commonly known as “El Yunque” is part of th e Caribbean National Forest located on the Northeastern block. Sierra de Luquillo’s highest peak is El Toro and its elevation is close to 915 meters. The Sierra de Cayey is also a mountain range in the island and its average elevation of 122 meters found in the Ce ntral block. There are many geological formations in Puerto Rico, San Sebastian, Lares, Cibao, Aguada, Aymamon and San Juan Formation. The oldest, San Sebastian form ations on the island, are composed of limestone.


7 Table 1. Soil type and soil se ries at study sites (NRCS, 2003) Site Soil Type (Order) Soil Series Common Clay Mineralogy BONUS Area Station 1-8 Mollisols Vertisols Colinas clay loam, 20 to 60 percent slope, eroded Mabi clay, 5 to 12 percent slope, eroded carbonatic mixed Vieques Mosquito Bay St. 1-2 Inceptisols Tidal Swamp (ocean sediments) Vieques Loam 5 to 12 percent slope, eroded mixed sandy Kiani Lagoon Inceptisols Vertisols Tidal Swamp (ocean sediments) Vieques Loam 5 to 12 percent slope, eroded Ponceno clay mixed sandy mixed Barracuda Lagoon Vertisols Tidal flats (mangrove area) Fraternity clay mixed smectite North and South Jobalos Lagoon Inceptisol Tidal Swamp (ocean sediments) Tidal flat Descalabrado clay loam, 20 to 40 percent slope, eroded mixed mixed (saprolite, silt) Operational Post Inceptisols Descalabrado clay loam, 20 to 60 percent slope, eroded Fraternity clay mixed (saprolite, silt, hard rock) smectite Caribbean National Forest (CNF) Cocoa Beach Tidal Swamp (ocean sediments) mixed Espiritu Santo River Tidal Swamp (ocean sediments) mixed El Verde Experimental Station Ultisols Yunque cobbly clay, 40 to 80 percent slope, extremely rocky kaolinite


8 A wide variety of soils series, spec ifically 215, are f ound in the diverse topography of the island. Table 1 summarizes all soil series and th e soil types found in the study areas. Since soil formation is driven by many factors, which include topography, climate, parental ma terial, biological factors an d time, it is reasonable to have such a diverse soil composition in an island whose origin is volcanic and where micro-climates dominate various ecosystems. For example, the Sierra de Luquillo dominates the topography of the Northeastern block in which more than two hundred inches of rain are received per year. Ther efore the most abundant soil types are those formed by high humidity and temperature. El Yunque Cobbly clay 40-80 extremely rocky (this is unclear) domina tes the region where samples were taken. The El Yunque series consisted of deep a nd well-drained soils on slopes. They are the result of weathering from andesitic to basaltic marine deposits. The deposit from which the series is formed comes from the Tabonuco and Hato Puerco Formation (NRCS, 2003). Average temperature in the formation site is 20 C and average precip itation is 470 cm of rain per year. At the Caribbean National Forest (CNF), “El Yunque”, the Humatas, Zarzal, Coloso, Cristal, Luquillo and Sonadora soil s series dominate the mountain ecosystems, most of them resulting from the weathering of rocks. The other factor that should be considered in the rainforest is the influen ce of organic matter from leaf litter, which makes the soils acidic. At the western area of the island, there are more than a hundred soil types. The area of Rincon where the power plant was cons tructed in the early 1960’s is dominated by limestone and epiclastic outcrops. The soils that dominate the study areas are the


9 Colinas and Mabi clays. The Colinas soil seri es consists of well-drained soils that are calcareous and moderately permeable. Colin as clays are the result of limestone weathering in a humid to sub-humid climat e. The cation exchange capacity of the Colinas soil series has been estimated to be 25-37 meq/100 g for the su rface layer of soil, 0-25 cm, and the soil has a pH range of 7.9 to 8.4 (NRCS, 2003). The second dominant soil series at the Boiling Nu clear Superheater (BONUS) reactor site is the Mabi soil series. Mabi soils are deep, poorly draine d and have poor permeability. They formed from volcanic material in a climate that ra nges between 25-26.1 C and 178 to 203 cm of rain per year. Cation exchange capacity of the Mabi soil series has been estimated to be 25-37 meq/100 g for the surface layer of soil, 0-20 cm, and has a pH range of 4.5 to 6.5 (NRCS, 2003). At the island of Vieques, soil production is the result of the island’s bedrock. The bedrocks are mainly from marine deposits of limestone and volcanic lava among other materials (ATSDR, 2002). Five soil series are found in the study areas in Vieques: Vieques, Poncena, Fraternidad, Descalabrado an d Tidal flat. The Vieques soil series is best described as moderately deep and well-drained with rapid permeability (NRCS, 2003). The Vieques soil series results from th e weathering of granitic rocks and they are commonly in dry ecosystems. Annual precip itation is about 89 cm per year and the average temperature is about 25.5 C. Vieque s soil series cation exchange capacity is 2535 meq/100g at the surface layer (0-20 cm) and the pH range is 6.1-7.8. The Poncena series is a deep soil that is moderately well drained, calcareous and has poor permeability. The series forms from fine-textured sediments derived from volcanic rock and limestone. The temper ature remains around 26.1 C, and the annual


10 precipitation is 89-127 cm per year. Its ca tion exchange capacity is 40-55 meq/100g at the surface layer (4 cm) and the pH range is 6.6 to 8.6. Climate fo r the soil series of Descalabrado and Fraternidad is similar to that discussed earlier where the around temperature is 26.1 C and pr ecipitation between 76-89 cm. They were formed from volcanic rock and limestone with respectiv e cation exchange capacities between 30-50 meq/100g and 35-55 meq/100g. Due to all the precipitati on received in the mountainous portions of Puerto Rico we know that the stability of the terrain is also important. Ther efore many scientists have used the island as living laboratories to study (landslide) mass movement (Larsen and Torres-Sanchez, 1998, Larsen and Par k, 1997, Larsen and Simon, 1993 and Scatena and Larsen, 1991). Landslides are the main s ource of erosion and sedimentation of the watersheds. This phenomenon has been followe d closely in the CNF where Larsen and Torres-Sanchez, (1998), Larsen and Park, (1997) and Larsen and Simon, (1993) have investigated how and where landslides occur. Sediments resulting from landslides reach streams, rivers and reservoirs, and eventual ly the ocean. Areas such as the CNF are vulnerable to this type of event. Therefore, watersheds in vicinities such as Espiritu Santo Estuary, as well as Rio Blanco and Sonadora among many others, receive a variable amount of sediment on a yearly ba sis. Hurricanes, tropi cal storms and human activity such as road construction all contri bute to landslides (Scaten a and Larsen, 1991). For instance, in 1989 during Hurricane Hugo ove r 400 landslides were registered at the northeastern mountains of Puer to Rico (Scatena and Larsen, 1991). Such events have not been replicated in recent times.


11 Hydrogeology of study sites Since all study sites occur in different ge ological settings inside the main island and Vieques, it is important to describe some of their hydrological characteristics. The CNF receives over 500 cm of rain per year. At the other end on the western coast in the municipality of Rincon, the av erage precipitation rate is 102 cm of rain per year. The island of Vieques is divided in two regions. In the western side of the island rainfall precipitation averages 127 cm of rain pe r year and the eastern portion receives approximately 64 cm of annual rainfall. Rain is the sole source of freshwater on the island and watershed formation is related to re gional precipitation. The CNF therefore is one of the main sources for freshwater in the island. More than 13 rivers are born there, producing water for the entire population on the ea stern region of Puerto Rico. In the municipality of Rincon near the BONUS nucl ear reactor, there are two major rivers, the Rio Grande de Aasco and Rio Loco. Vieques does not have any major rivers. Se veral aquifers are the main sources of water. The Esperanza valley is the primary aquifer, while Resolucion, Playa Grande and Camp Garcia are minor ones. These aquifers fo rm within alluvial de posits located in the low flat valley along the coasts. Rainfall is the major source for fresh water recharge of these aquifers, and Esperanza-Resolution Vall ey aquifers are made of fine grained alluvium from weathering of dioritic rocks. At the main island, aquifers are made of limestone, volcanic rock and alluvium. In the north coast aquifers consist of heterogeneous body of interbedded permeable and poorly permeable material. At the southern region, aquifers are made of allu vium and although the s outhern region is the


12 driest region of the island, aquifers in th is region produce more water than those on the northern coast. Studies on Cs-137 mobility Cs-137 is one of the most common anthropogenic radionuclides that have been found in clay-rich, especially illite, sedime nts (Livingston, 2000, Johnson-Pyrtle et al., 2000, Moon et al., 2003). Cs-137 has a positive ch arge and it requires a negative charge to form a bond and attach to soil and sedi ment particles (Arapis and Karandinos, 2004, Olsen, 1981). In the natural environment, bi nding of Cs-137 to particles preferentially occurs on the negatively charged surfaces of clay minerals. These surfaces are provided by the double layers of illite and smectite and the single layer of kaolinite (JohnsonPyrtle, 1999). On the island of Puerto Rico, the higher abundance of clay-sized particles may be attributed to the dissolution of carbonates and th e weathering of terrestrial minerals (Fox, R. 1982, Moon et al. 2003 a nd Pett-Ridge, et al. 2009). The lower abundance of some clay particles may reflect high accumulation rates of other sedimentary material including biogenic debr is (Moon et al., 2003). Other factors that affect th e distribution of radionuclides in aquatic and terrestrial sediments include particle mixing, cation-co mpetition reactions, salinity and sediment mineralogy (Santschi, 1989; Moon et al., 2003). It has been shown, for example, that the activity of Cs-137 varies as a function of mi neralogy, particle size a nd the salinity in the Siberian Arctic’s Laptev Sea-Lena River estuary re gion (Table 2).


13 Table 2. Laptev Sea-Lena River estuary re gion average sediment. Cs-137 activity (Bq kg-1) and grain sized distribution (Adapted from JohnsonPyrtle and Scott, 2001) Cation exchange reactions that often occur in fine-grained clay-rich sediments can result in radionuclide enrichment or depl etion (Fanning et al., 1981). For example uranium is present at higher concentrations in clay minerals than in limestone in Florida Platform (Fanning et al. 1981). The term cation exchange capacity refe rs to the ability of a particle to retain ions with positive ch arge (cations) in th eir negatively charged structure. Moon (2003) suggests that th e stratigraphic record of the upper centimeter of sediments from the Norwest Pacific Ocean is strongly influenced by particle mixing by benthic fauna. This mixing can result in anthropogenic radion uclide enrichment in surface sediments. Cs-137 can also be bioturbated by the activity of plants in the soil. This phenomena by which Cs-137 is absorbed or removed from soil is called the plant-tosoil transfer (Tyler et al. 2001) Cs-137 mobility in soil and sediment is re latively slow and a factor affecting the slow mobility is the irreversible sorption pro cess in soil particles (Arapis and Karandinos, 2004). In sandy textures, Cs-137 adsorption is low due the high permeability of the minerals (Olsen, 1981). However, in soil and sediment containing high quantities of fineSalinity Cs-137 activity (Bq/g) %Silt + % Clay (< 63 m) %Clay (<4 m) Marine average values 27.18 6.00 62.63 18.95 Lena River estuary average values 25.65 11.22 83.25 27.06 Study region average values 25.62 7.08 66.99 19.01


14 grained material, the likelihood of adsorpti on of Cs-137 to particles is much higher (Johnson-Pyrtle, 1999). Slow mobili ty in porewater is due main ly to the fact that Cs-137 has to compete with seawater cations, esp ecially potassium (Olsen, 1981). This should decrease sorption and therefore increase m obility. Mobility of Cs-137 in freshwater ecosystems such as rivers is related to th e interactions of the radionuclide with the suspended particulate matter. It also may depend on pH, the time needed for equilibrium and the nature of the suspended particle in the water among others factors (Ciffroy et al. 2009) Sources of radiation in the environment Environmental radiation comes from two sources; background and man-made (NRC, 2006). Background radia tion is the result of cosmogeni c and terrestrial radiation, while man-made radiation is of anthr opogenic origin. Some common terrestrial radionuclides belong to the decay series of U-238, Th-232 and U-235. Due to Uranium’s long half-life (4.47 billion year s), its decay products (Th-234, U-234, Ra-226 and Rn-220 among others) persist in terres trial and aquatic environments They are commonly found in rocks, minerals and groundwater. Some decay products, such as Rn-222 and Ra-226, have become radiation sources to humans as they can be ingested or inhaled after their production (Zikovsky, 2006). Besides volcanoes, rocks also contribu te to the naturally occurring global background. U-238 decay products, K-40, Ra-226, Th-232 and Sr-90 are sources of radiation present in rock. Example locations where these radionuclides contribute to the


15 background levels in soils are Egypt, India a nd Germany. (Ahmed et al. 2006; Sadasivian et al. 2006; Takeda et al, 2006; Anoruo et al 2002; Farai et al. 2001; Jibiri, 2001). Nuclear weapon testing Many of the studies conduc ted after the initial nuclear weapons testing have focused on understanding the fate and transpor t of radionuclides in aquatic ecosystems. When investigating their fate and transport in the environment, researchers have taken into consideration that once deposited in the ecosystem daughter radionuclides can be present as the decay product of the original radioisotope. Researchers have also focused on the effects of radionuclides on local food web dynamics and their associated health implications. Above ground testing has contributed to th e release of radionuclides into the environment via fallout. The characteristics of fallout particles depend on the height of the burst and explosion yield. Once in the atmosphere fallout can occur in two types, dry deposition or wet deposition. Since particle s introduced into the atmosphere display various characteristics, their i ndividuality is important durin g the fallout process. For example, at higher altitudes dus t particles may play a role in the condensation of rain drops and this process is re sponsible for the removal of the smaller particles in the atmosphere. Particles exceeding th e 10m with a density of 2.5 g/cm-3 generally fall out within a few hours at sea level. The fate of smaller particles in the troposphere will depend on laws of dispersion. Smaller particle s can penetrate the stratosphere. It can take months to years before these particles reach terrestrial and aquatic ecosystems (Beck and Bennett, 2002, Eisenbud and Gessel, 1997). The main parameters controlling the


16 deposition of radioactive partic les in the stratosphere and tr oposphere are the temperature and precipitation. The warmer air in the tr opics will allow the particles to reach higher in the atmosphere. Once temperature decrea ses and precipitation occurs, radioactive particles will be deposited in the ecosystem. Underground testing has also been broadly conducted. The number of underground nuclear weapons tests in the US alone exceeds 800, a number that surpasses the combined number of above ground tests conducted by Russia, China, France and the United Kingdom (Beck and Bennett, 2002). As a result of nuclear weapons testing more than 40 radionuclides were released to the environment. Some of the most common of these radionuclides are I-131, Cs-137, Sr90, Te-132 and Pu-239-240 (Simon et al. 2004, Beck and Bennett, 2002). Environmental risk caused by anthropogenic radionuclides in the Caribbean Region Twenty years following the introduction of Cs-137 into the atmosphere from the Chernobyl accident, little research has been conducted in the Caribbean region to assess the effects of radionuclides. Cuba, Venezuel a, Brazil and Costa Ri ca are countries where radiogeochemical studies have been conducte d. Knowing the response of the ecosystem to Cs-137 is important since it has variable depositional behavior. Zhiyanski et al. (2005) stated that Cs-137 mobility in the tropical rainfo rest of Bulgaria is affected by litter. This serves as a biochemical agent that enables mobility of Cs-137 and availability for plant uptake. Since the mobility of the radionuclide in the soil depends upon soil conditions, pH, soil series, solid /liqui d distribution coefficient and organic matter content among others, plant uptake in the acidic forest soil will affect the mo bility of the radionuclide.


17 Zhiyanski’s findings were confirmed by other researchers investigating the Caribbean region. Most of the studies conducte d were on the soil to pl ant transfer in the tropical forest of Venezuela and Costa Ri ca (LaBrecque and Cordoves, 2004, Bossew and Strebl, 2001). LaBrecque and Cordoves, ( 2004) found activities levels ranging from 2 Bq/kg to 14.8 Bq/kg in surface soil samples at the Gran Sabana and Sierra Pacarima in Venezuela, and about 15% of the samples had values above 5Bq/kg. These activity levels are similar to those reported by Bossew and Strebl (2001) in the tropical forest of Costa Rica. LaBrecque and Cordoves, (2004) reported that Cs-137 activity was influenced by local rainfall. Rainfall in the region of Gran Sabana and Sierra de Pacarima is very similar to the rainfall re ceived at the Caribbean National Forest, El Yunque, which is about 200-500 cm of rain per year. LaBrecque and Cordoves (2004) and Bossew and Strebl (2001) proposed a direct relationship between rainfall (precipitation) and the repor ted Cs-137 activities. AlonsoHernandez et al. (2006) also report ed a similar positive relationship (r2=0.92) showing an increase in levels of Cs-137 with an increa se in wet deposition (ra infall) at Cienfuegos, Cuba. Other studies in the Caribbean region re late elevation and soil chemistry as well as precipitation to higher ac tivities of Cs-137. That is the case of for Venezuelan organicrich soil and dry deposition by clouds at hi gh altitudes (LaBrecque and Cordova 2004). Alonso-Hernandez (2006) found that grain si ze also has a direct relationship on the activity of Cs-137 in Cuba.


18 Health risk associated with radiation Health risks most often associated with exposure to radioactivity include cancer, developmental effects (e.g. mental retarda tion), and non-genetic effects (UNSCEAR, 1982). Some types of cancer related to radi onuclides exposure are skin cancer, leukemia and lung cancer. Ionizing radiation exposur e is a known, and well-quantified, cancer risk factor. Estimation of cancer risk following radi ation exposure is very uncertain with respect to regulatory and/or popular concerns. One reason is that risk estimates are usually applied to populations that are different from the populations on which the estimates are based. Figure 2. Assumptions for the extrapolation of the cancer risk vs. radiation dose to low-dose levels, given a known risk at a high dose (Alonso-Hernandez 2006). Alonso Hernandez et al. (2002) measur ed the dose response of Po-210 and Cs-137 as the result of fish, mollusk and crustacean consumption at Cienfuegos Bay, Cuba. Observations of anthropogenic radionucl ides in fish promote an understanding of how of food web dynamics influences the introduction of these man-made nuclides. Linear Res p onse Threshold Dose


19 Alonso-Hernandez et al. (2002) concluded that the committed effective doses from ingestion of seafood containing Cs-137 are negligible. However, activities (50–125 Bq/kg w.w.) found in fish and mollusks from Cienfuegos, Cuba, actually exceed recommended levels for human consum ption reported by the UNSCEAR (2000). The UNSCEAR as well as the Internati onal Atomic Energy Agency (IAEA) and the Nuclear Regulatory Commission (NRC) have cr eated limits based on the hypothetical data of responses from the exposed population afte r the Chernobyl accident. The no-threshold hypothesis means that increased effects are obser ved at high doses of radiation, but effects at low doses are not known. Therefore ther e is no way to determine at which dose the cancer incidence will increase in the population (Figure 2). The As Low As Reasonably Achievable (ALARA) Principal should be followed at all times to avoid exposure to radiation. As documented in the past, dos es that exceed (25-50 rems) can become detrimental to human health.


20 STUDY AREA Boiling Nuclear Superheated Reactor During the early 1960’s the US Atomic Energy Commission (US AEC) and the Puerto Rico Water Resource Authority (P RWRA) constructed the Boiling Nuclear Superheated (BONUS) reactor facility in orde r to investigate the technical and economic feasibility of the integral boiling-superheat ed concept (US DOE, 2003) (Figure 3). The prototype power plant was constructed in a coastal lowland area on the western coast of Rincon, Puerto Rico. The Boiling Nuclear Superheated (BONUS) reactor experienced numerous problems, including 106 unintenti onal reactor shutdowns, one of which resulted in the release of a radionuclid es mixture which concentration was 582.3MBq into the atmosphere (US DOE, 2003).


21 Figure 3. BONUS Reactor sampling area The BONUS reactor facility was operationa l for a total of six years (1962-1968) before being decommissioned, a process whic h was completed in 1970 (US DOE, 2003). In 1993 the basement of the entombed reacto r was flooded by rainwater as a result of a failed exhaust fan (US DOE, 2003) which may ha ve resulted in furthe r contamination if radionuclides were present. Recently, the local government proposed to develop a museum at the site of the BONUS facility (US DOE, 2003). Atlantic Fleet Weapons Training Facility The island of Vieques is located 11km s outheast of the main Caribbean island of Puerto Rico (Figure 4). Vieque s is 32 km long and 7.2 km at its widest point. The island is currently inhabited and its population is close to 9,000 people. For more than six


22 decades it was used as the Atlantic Fleet Weapon Training Facility (AFWTF) (US EPA, 2001). During the time that the US Navy used the island for military purposes, the population decreased, about 26% (Ayala Ca rro, 1999). The population changed from about 10,000 in the 1940’s to 7, 800 in the 1970 ’s and recently the number went back to 9,000 people. From 1999 to 2003 ammunitions containing de pleted uranium were released into the atmosphere as the result of military practices, and pollutants made their way into the local aquatic and terrestria l ecosystem (ATSDR, 2002; Li ndsay-Poland, 2001; Yarrow, 2000). The US Navy targeting exercises at the AFWTF ceased in 2003 and unexploded ammunitions, shell casings, scrap iron, and other military debris, including depleted uranium ammunitions are still present at the AFWTF (ATSDR, 2002; US EPA, 2001; US Navy, 2002). Figure 4. Vieques sampling site


23 The pollution released into the atmosphere by military practices has traveled more than 7.9 miles reaching the residential area of Vieques, and radionuclides may have traveled that distance as we ll (ATSDR, 2002). Pollutant concentrations were measured by the Agency for Toxic Substances and Dise ase Registry based on the information that the ammunitions used in Vieques containe d TNT, trace amounts of hydrogen cyanide, aluminum and ammonia among others. It was found pollutant concentrations were below the limits permitted by regulatory agencies. Other studies conducted on Vieques indicate that copper, lea d, nickel, cadmium and manganese are present in high concentratio ns in plants, crops and vegetables (Diaz and Massol-Deya, 2003). Porter (1999) conducte d the sole study of current radiological conditions on the island of Vieques. Caribbean National Forest El Verde Experimental Station is the ma in site of a series of experiments conducted by the Department of Energy ( DOE) on the island of Puerto Rico in 1965 (Figure 5) (NRC, 2001; PRNC, 1970 ). The purpose of these series of experiments by the DOE was to understand the behavior of radi onuclides in a tropical ecosystem. These experiments entailed the inocul ation and irradiation of trees in El Verde (PRNC, 1970). The radionuclides used for the irradiati on and inoculation we re Cs-137, Cs-134 and tritium (NRC, 2001; PRNC, 1970). Reptiles, am phibians and insects were also dosed with radionuclides to study toxic effects in the tropical fauna (NRC, 2001; PRNC, 1970).


24 Figure 5. Sampling sites at Espritu Sa nto Estuary, Rio Grande, Puerto Rico. Irradiated Cs-137 was used as a gamma source for radiation experiments on selected vegetation plots (PRN C, 1970). As the results of this experiment, radiation levels in the area increased from 5 mrem/h r to 200 mrem/hr (US NRC, 2001) which is a forty fold increase of the radiation level in th e area prior to the experiment. It was also estimated that approximately 50 % of the ces ium adsorbed by trees was eliminated. Several trees were inoculated with 777 MBq of tritium at the study site by injecting their trunks with this radionuclide at an angle of 45 and a height of 25 cm above ground to analyze transpiration rates (PRN C, 1970). Reports from the Puerto Rico Nuclear Center (1968) indicated that once rel eased into the environment ra dionuclides such as Cs-134 and Sr-85 were bound in the litter and to the su rface soil of the rainforest. This fact was verified by the rapid increase in Cs-134 and S r-85 activity in the first 5 cm of the soil where the experiment was conducted (PRNC, 1968).


25 HYPOTHESES The introduction of anthropoge nic radionuclides in Puerto Rico which occurred as the result of governmental, military and indus trial activities has resulted in local distribution of Cs-137 in Puerto Rico. The distribution is expe cted to vary from site to site. In order to assess the impact of such activities, this proposed work will test the following hypotheses: 1. Cs-137 concentration found in the tropical rainforest watersheds and estuaries will have a down slope gradient and the hill area’s higher Cs-137 concentrations will be located near Cocoa Beach. 2. Cs-137 concentration in the island is the result of global fallout and not the result of governmental activities. 3. Mineralogy of the study area will be similar in the three different ecosystems due to geological formation of the island. 4. Mobility of Cs-137 will be determin ed by the abundance of clays and organic matter in the different ecosystems. 5. There is no difference between the re tention of Cs-137 by soil and sediment samples at the study areas 6. Soil and sediments from study areas have the capacity to retain heavy metals and other chemical elements


26 OBJECTIVES The overall goal of this study is to de scribe the behavior and determine the distribution of radionuclides that have been introduced into the environment of Puerto Rico. To meet this goal the fo llowing objectives were pursued: Quantify gamma activity levels at each study site. Examine the fate and transport of an thropogenic radionuclides in the soil and sediments of the three study areas. Conduct a grain analysis at each site to quantify the am ounts of clay available for anthropogenic radionuclides. Assess the chemical composition of sediment and soil samples.




28 INTRODUCTION Samples were collected from three municipa lities in the island of Puerto Rico in order to investigate their various radioge ochemical characteristics. The island of Vieques, El Verde Experiment al Station, along with the Espiritu Santo Estuary and the adjacent area of the BONUS f acility are described below. Vieques Vieques is the largest of the adjacent is land municipalities of Puerto Rico. Located in the Caribbean Region, the social history of Vieques’ is as rich as its environment. Vieques has approximately 115 km2 of coastline which encompasses various ecosystems. From the dry coast to a cen tral ridge that reaches its highest point at Monte Pirata located in the southwest region. The elevation of Vieques varies approximately 300 m (Renken et al. 2000). Su rface freshwater on the island is provided by rainfall which contributes an average of 200 cm of rain per year (Renken et al. 2000). The rain serves as a freshwater source to local rivers and streams that flow intermittently in direct response to rain ev ents (Renken, et al. 2000). Vieques was formed from igneous and vol canic rock, mostly granodiorite, quartz diorite, and some lavas, which created the be drock of the island. This bedrock is exposed and weathered on most of the western half of the island and some portions of the eastern half of the island. The island’s geologic formations include two alluvial valleys,


29 Esperanza and Resolucion (Renken et al. 2000). The Esperanza alluvial valley is about 0.5 km to 1 km wide and 5 to 6 km long and the alluvial deposits are about 18 m thick (Renken et al. 2000). The alluvial sediment ary deposits generally consist of a selfcontained mixture of gravel, sand, silt, and clay (Renken et al. 2000). In 1989, the United States Geological Survey (USGS) reported that groundwater flow in Esperanza valley alluvial deposits was toward the south and toward the sea. The Resolucion alluvial valley is, on average, 9 m thick and overlies the bedr ock, which is composed of granodiorite and quartz diorite. This valley also has a semi-c onfining clay layer at about 6 to 9 m below ground surface. Resolucion, which is located ne xt to Monte Pirata, receives more rainfall recharge than Esperan za (Renken, et al. 2000). For more than 60 years, the eastern porti on of the island of Vieques served as a weapons testing site for the U.S. Department of Defense. During the 1990’s, U.S. Navy targeting exercises at the AFWTF involved the use of armor-piercing ammunitions containing depleted uranium (Yarrow, 2000). Im pact studies indicate that anthropogenic material released into Puerto Rico’s environment resulted in the contamination of natural resources via surface and groundwater tran sport (Diaz and Massol-Dey, 2003; U.S. EPA, 2001; Hunter and Arbona, 1995). Boiling Nuclear Superheated Power Station (BONUS) The BONUS superheated nuclear power plan t prototype facility was constructed to investigate the technical and economic feasibility of the integral boiling-superheated concept. BONUS was the eighth nuclear power plant constructed in the U.S. The main components for the Center for Energy and Envi ronmental Research in Puerto Rico were


30 the BONUS facility and three other sites. This complex consisted of 176 acres managed by the University of Puerto Rico and the Pu erto Rico Electric Power Authority. The US Atomic Energy Commission (US AEC) and the Puerto Rico Water Resource Authority (PRWRA) began construction of th e BONUS Facility in 1962. The BONUS facility was constructed w ithin the coastal lowland area on the western coast, near Rincn. BONUS was cons tructed less than a mile from the merging Caribbean Sea and Atlantic Ocean. The beach next to the facility is considered to be one of the islands’ best surfing beaches and is frequented by many surfers. No nearby river or stream is adjacent to the BONUS facilit y. As such, surface water-mediated soil and sediment transport depends on local rain events, which are generally only a few centimeters per year. During its operation (a total of si x years 1962-1968), the BONUS reactor experienced 106 unintentional reactor shutdowns As of yet, there are no conclusive studies to assess the total environmental impact of this facility. Records indicate that during one unintentional shutdown, the B ONUS reactor released 582.3MBq into the atmosphere and a radiation dose between 180110 mrem was received by two employees at the BONUS facility (US DOE, 2005). A va riety of radionuclides were used in the daily operation of the prot otype power plant. Major efforts were made to remove all ra dioactive material associated with the BONUS facility during the decommissioning pro cess. During this process, the reactor and radioactive material were entombed with in a concrete monolith (Table 3). Recent reports indicate that radioactive material remained in the BONUS facility after the decommission process was completed (Table 4). After decommissioning the plant,


31 approximately 4.81x108 Bq of radioactivity was containe d in the pipes, as well as other external components to th e power plant (US DOE, 2005). Approximately 1.96 1015 Bq was left inside the entombment system with the expectation that the total activity would decrease as time passed as a result of ch aracteristic decay processes of individual radionuclides (US DOE, 2005). Table 3. Past activities (measured in 1968) a nd recent activity estimates (determined in 2001) of principal radionuclides entombed with the BONUS reactor for 1968 and 2001 (Adapted from US DOE, 2005). Radionuclide Half-Life Activity (Bq) 1968 2010 Cobalt-57 271 days 8.247 x 1013 0 Cobalt-60 5.27 years 5.764 x 1014 2.300 x 1012 Nickel-63 96 years 3.108 x 1013 2.323 x 1013 Manganese-54 312 days 3.785 x 1013 0 Iron-55 2.7 years 1.242 x 1015 2.903 x1010 TOTAL 1.971 x 1015 Table 4. Past activities (measured in 1968) a nd recent activity estimates (determined in 2001) of principal radionuclides external to the entombed BONUS reactor for 1968 and 2001 (Adapted from US DOE, 2005). Radionuclide Half-Life Activity (Bq) 1968 2010 Manganese-54 312 days 4.070 x 106 Cobalt-60 5.27 years 3.7 x 108 1.477 x 106 Zinc-65 244 days 5.92 x 107 Silver-110m 250 days 3.11 x 105 Antimony-125 2.77 years 1.41 x 106 Cesium-137 30 years 5.5 x 108 2.088 x 107 TOTAL 4.81 x 107 During the period of operation Cs-137 was the primary radionuclide present at the BONUS facility. Ni-63 was also found at the fac ility when the reactor was being utilized. Ni-63 has been identified as the radionuclide of greatest concern, primarily because of its 96 years half-life. Ni-63 is the main contribut or to radiation in the entombed system (that


32 was buried in the facility) at the BONUS f acility. However, Cs-137, was the greatest contributor to the total activity of the entomb ed system‘s external pipes and other various structural components at the BONUS facility. During hurricane George in 1998, the enclosed dome was flooded and the basement doors leaked its contents (US DOE 2005). Debris from the plant was then carried down slope through storm water runoff (US DOE, 2005). It is important to understa nd the geochemical characteri stics of soils surrounding the BONUS facility because of the existing pot ential for leakage at the BONUS facility, as well as the previous introduction of mate rial from this prototype power plant to the local environment. Given the history of the BONU S facility and the potential release of radionuclides into the local environment, geochemical properties of soils surrounding the BONUS Facility should be determined. Previous researchers have iden tified the erosion of radionuclides as a secondary source of sedi ment contamination (Claval et al. 2004 and Charmasson, 2003). Radiogeochemical prope rties of the nearby Dome Beach depositional system were examined in order to investigate the potential for local Cs-137 deposition and removal, as well as to identi fy the potential for BONUS-derived material to be incorporated in to the local food web. El Verde Experimental Statio n and Espiritu Santo Estuary The Caribbean National Forest (CNF) is th e only tropical rain forest in the United States National Forest System. This unique ecosystem has been protected since 1876 when it was declared the firs t natural reservoir of the Ca ribbean region. In 1903, the US


33 National Forest System began managing this forest of 5,115 acres. Today the CNF is composed of more than 28,000 acres and its boundaries are sh ared by thirteen municipalities on the east coast of the island of Puerto Rico. The CNF environment is characterized by its year round tropic al climate and the large amount of rainfall. It has been calculated that precip itation at the CNF exceeds 508 cm of rain per year. Trade winds bring all th e clouds to the highest po int of the rainforest where the highest elevation is 1,076.9 m causi ng condensation and fu rther precipitation which will increase riverine input to local ecosystem. Stream flows in the rainforest are highly variable ( ten-fold increases in discharge have been recorded within an hour) and fluctuate with rainfall (Larsen and TorresSanchez, 1998) Within this tropical rain forest, streams are often located in steep, bedrockand boulder-lined channels within narrow valleys. Headwater channels, frequently lined with trees, have shallo w pools (<0.75 1.0 m deep) and exhibit relatively constant temperatures ranging throughout the year between 18 and 24 C (Larsen and Simon 1993). Second and third-order tributarie s present in CNF are typically shallow (< 1 m deep) and have relatively open canopies (Larsen and Torres-Sanchez, 1998). The pH of the rivers in the rainfo rest ranges between 7 and 7.7 w ith an average pH of 7.2. Because of its specialized ecosystem the CNF has often served as a living laboratory for numerous scientific investigations. CNF also serves as a model for many other reserves worldwide. One of the bigge st experiments that occurred at the CNF was performed during the nuclear testing era. The El Verd e Research Station, which encompasses 156 acres of the CNF, served as a site for local terres trial ecology radiation experiments that were managed by the US Atomic Energy Commission (AEC) from 1964


34 to 1976. These experiments were undertaken 42 4 m above sea level as part of the Rain Forest Project (RFP) of the Puerto Rico Nuclear Center, (Odu m and Drewery, 1970). Researchers involved with the RFP conducted a wide array of expe riments designed to investigate the rainforest’s response to radi ation. These experiments examined a variety of processes, including mineral cycling a nd water movement. The primary experiment conducted at the El Verde site involved the i rradiation of a forest parcel with 3.7 x 1014 Bq Cs-137 source. The US Department of Agriculture compar ed the results of th is study with those obtained from similar experiments within the Panamanian forest (Odum and Drewery, 1970). Forest metabolism and water excha nge studies were also conducted. Major objectives of this initial and subsequent smaller-scale gamma ir radiation experiments conducted at the study site were to: 1) dete rmine the radiation effects on a rainforest ecosystem (technically classified as a subtropi cal wet forest), 2) examine the cycling of fallout elements, and 3) gain additional unde rstanding of vertical and horizontal forest structure and ecological proce sses including nutrient cycling, energy flow, and forest regeneration. The Rainforest Project Researchers conducted metabolism and defoliation experiments using various radionuclides in order to observe measurable ch anges in the forest vegetation (Table 5). The radiation source of 1014 Bq of Cs-137 (close source) was installed in December, 1964. The period of irradiation was Janua ry 19, 1965 to April 26, 1965 (Desmarais and Helmuth, 1970). Desmarais and Helmuth st ated that “damage around the radiation


35 source was quite obvious by the end of the radiation period and persisted throughout the following year”. They also reported that the damage was confined within an area of 12 m radius of the radiation source.


36 Table 5. Summary of radionuclides us ed at the Rainforest Project. Isotope Date Applied Original Activity (Bq) Half-life 3H Jan. 6 1967 74 x 107 12 years 3H 32P 32P May 1968 May 1969 May 1972 18.5 x1083.7 x 107 17.02 x 108 12 years 14.5 days 14.5 days 137 Cs 86 Rb 85 Sr 54 Mn Sept. 18, 1968 Sept. 18, 1968 Sept. 18, 1968 Sept. 18, 1968 17.02 x 10665.453 x 107 7.03 x 106 12.58 x 106 30 years 45 days 64 days 313 days 85 Sr 134Cs Aug. 10, 1967 Aug. 10, 1967 3.7 x 10729.6 x 106 64 days 2.062 years 65Zn Aug. 3 1967 11.1 x 107 244 days 85 Sr 134Cs 54 Mn Jan. 6, 1966 Jan. 6, 1966 Jan. 6, 1966 3.7 x 1073.7 x 107 3.7 x 107 64 days 2.062 years 313 days The local destruction of fo rest structure resulted in new patterns of succession (Odum and Drewery, 1970). The vegetation struct ure of a tropical rainforest is measured by the number and quantity of species found, and form and position of plants within the forest. Of great importance for the RFP was the development of the ability to understand the succession pattern. In November 1965, the radiation source area was re-sampled and it was determined that the area that was im pacted by the radiation had grown from the original 12 m radius to a radius of 24 m. Defoliation had taken place in the damaged area and the optical density had declined (less l eaves will absorb less light) (Odum & Drewry, 1970). The irradiation of the ecosystem with Cs -137 resulted in a notable increase in the


37 number of crownless plants w ithin a24 m radius and almost a complete loss of leaves from the local vegetation (Odum & Drewry, 1970). Table 6. Radionuclide movements on vegeta tion after their inje ction (Activity in Bq/g dry weight) (From: PRNC, 1985) Sample Cs-137 on Matayba sp Rb-86 on Dacroydes sp. Sr-85 on Dacroydes sp. 20 days 132 days 20 days 132 days 20 days 75 days Leaves 2.405 15.059 84.027 562.141 0.22 0.444 Twigs 0 18.796 0 2409.292 0.873 Wood, 1 ft above injection hole 206.053 17.168 739.667 536.352 2.346 Wood, at level of injection hole 1.702 26.042 967.254 507.677 0 Wood, base of tree 219.373 13.801 3255.63 1568.911 0.0 0.081 Bark, base of tree 6304.134 229.585 23178.65 Below detection limit 125.245 Water movement in soils was further examined using 74 x107 Bq of tritium in the form of titrated water. The tritium was applied to a soil parcel of 0.94 m2. Results of this experiment indicated that although most of the tritium passed thr ough the upper 18 cm of the soil as a pulse, a fraction remained in th e clay-rich soil. Another component of the water movement experiment involved injecting tritiated water into trees, via holes bored into tree trunks near the gr ound (Kline and Jordan, 1970). Leav es from the injected trees were collected and analyzed for tritium. Cs-137, Sr-85, Rb-86 and Mn-54 injecti on experiments on trees were conducted (Table 6) in order to investigate the effect s of radioactive fallout in tropical ecosystems (PRNC, 1970). These ‘fallout’ radiati on experiments, conducted from 1966 through


38 1970, included studies on insect and amphibi an ecologies, food web dynamics and effects of rainfall on the fate of radionuclides (d issolved and particul ate) (PRNC, 1970). The diverse aquatic and terrestrial ecosyste ms of the island of Puerto Rico and its municipalities have been studied by many res earches over the past decades. Cs-137 was locally introduced at two of the three study sites that were examined during this investigation. One of the major goals of this investigation is to determine the fate of the Cs-137 that was utilized during the previously descri bed experimental ac tivities. Another goal of this study is to document the reten tion and post-depositional behavior of global fallout Cs-137 in Puerto Rico’s various ecosystems.


39 METHODS Soil and sediment samples were collect ed from the BONUS facility, CNF, Espiritu Santo River Estuary and Vieques usi ng a polycarbonate core liner. At eight cores were collected at the BONUS facility. These cores were collected, in pairs, from each side of the fenced facility. Samples from the CNF were collected from the Radiation Center (used by Odum during the Rainforest Project ) and inside the protected area where tree injection experiments took pl ace. At Espiritu Santo River Estuary, samples were collected from mangrove areas, as well as from a low energy beach and along the mouth of the river. Surface sedi ment and soil samples, ranging 1-5 cm in depth, were collected from six sites (Ope rational Posting Lagoon, Kiani Marsh Lagoon, North and South Jobalos Lagoon, Baracuda Lagoon and Green Bay Lagoon) within restricted areas of the Vieques National Fi sh and Wildlife Refuge (VNFWR). These restricted areas at the VNFWR are not open fo r public use as they represent a potential threat to public health and sa fety, due to the potential exis tence of unexploded ordinance.


40 Figure 6. Sampling sites in Puerto Rico Figure 7. Surface and downcore Sampling Sites in Vieques, PR.


41 Two types of samples were taken duri ng this project, surface samples and core samples. Surface samples (1-5 cm deep) were taken using a shovel. Core samples were taken using the polycarbonate core liner (3 m long, 5 cm radius). Once collected, cores were sectioned using an extruder. The extr uder was built to allow for 0.5 cm and 1 cm sampling intervals. Core samples within the uppermost 10 cm were sectioned at 0.5 cm thickness. Samples from 10 cm in depth a nd beyond were sectioned utilizing 1 cm thickness intervals. Samples were cut usi ng polycarbonate spatulas which were rinsed with water between each sample to avoid cross contamination. The sectioned samples were stored in commercial brand zip-lock bags. The stored samples were transported to the USF College of Marine Science Aquatic Radiogeochemistry Laboratory in portable c oolers and upon arrival placed in -4 C commercial freezer. In preparation for laborator y analysis, samples were allowed to thaw and then aliquots of each section were wei ghed, dried, and reweighed to determine the sediment porosity. The samples were dr ied using a freeze drier at -20 C for approximately 72 hours.


42 Table 7. Sampling location, description of hab itat conditions. Mangrove Forest indicated by MF. Lagoon indicated by L. Marsh indicated by M Sample Site Location Type of Sample Type of Environment Water Depth (cm) Mosq.Bay Vieques Sediment MF 30 cm Kiani Lagoon Vieques Sediment MF 15 cm Green Bay Vieques Sediment M 10 cm South Kiani Lagoon Vieques Sediment M 10 cm Baracuda Lagoon Vieques Soil L N/A Operation Posting Lagoon Vieques Soil L N/A S. Jobalos Vieques Sediment MF, L 10 cm N. Jobalos Vieques Sediment MF, L 10 cm


43 Gamma analysis for Cs-137 Sediment and soil samples were prepared for radionuclide analyses following techniques described in Johnson-Pyrtle and Scott (2001). The samples, packaged in plastic test tubes, were assayed for gamma emitters using two Canberra high purity germanium well gamma detectors, connected to a Canberra Genie multi-channel analyzer which records the gamma spectra in 4096 cha nnels. The two detectors were calibrated using U.S. National Institute of Standa rds and Technology 4357 Ocean Sediment multiline and Canberra Industries MGS5 sediment standards. The resulting spectra were analyzed fo r the Cs-137 specific energy peak at 661 keV, and the activity of each sample was th en determined. Detector performance was verified using a multiline liquid standard, a Peruvian soil standard, Columbia River sediment standard and Ocean sedime nt standard prepared by NIST. Cs-137 inventories for each co re were calculated as follows: I = k i ti Ci, Where I = Cs-137 inventory (Bq/m2) for each sediment core, Ci = Cs-137 (Bq/g) measured in each increment, i = average particle density (g/cm3), ti = the thickness of each increment (cm), k = 10,000, a constant for converting (Bq/cm2) to (Bq/m2)


44 Gamma analysis of Pb-210 and Ra-226 Sealed in plastic vials, Ra-226 and Pb -214 were allowed to come into secular equilibrium by waiting for a period of no less than 30 days prior to gamma analysis. This period corresponds to approxima tely 8 half-lives of Rn222, the immediate daughter of Ra-226. Pb-214 is a daughter product of Ra222 and a precursor to Pb-210. Therefore, the activity of Ra-226 can be determined by determining the activity of Pb-214,. By subtracting the Pb-214 (Ra-226) ac tivity from the Pb210 activity, the amount of unsupported excess Pb-210 was determine d. Samples were analyzed for gamma emitters for at least 24 hours using the low background, germanium well detector. Sample activities at 46 keV (Pb-210) and 351 keV (Pb-214) were determined. Energy and efficiency performance were calculated and factored in prior to determining the actual activities (see appendix G). Grain size analysis Particle size distributions we re determined for all sediment and soil samples using the traditional sieving method and a Micromeritics Saturn DigiSizer 5200 (a high resolution laser particle size analyzer) (Jackson, 1958; Micromeritics Instruments Corporation, 2003). Prior to grain size anal ysis, one gram of each freeze-dried sample was sieved using a 63 m mesh size screen, and a 5 % solution of Sodium Tripolyphosphate (Calgon). This process created a slurry which facilitated the wet sieve of the sample Particles larger than 63 m were co llected from the sieve, air dried and weighed. Finer-grain size material that passed through the 63 m sieve was weighed and grain size fractions were determined utilizing the Saturn DigiSizer. Particle size classes


45 were defined as follows: clay size 4 m; 4 m < silt 63 m; sand and larger > 63 m. The Saturn DigiSizer analysis provided a raw percentage of each particle class. This percentage was then subtracted from the data obtained from sieving and weighting of the sample to obtain a final class percentage. Loss on ignition analysis Organic matter content was determined using standard Loss on Ignition (LOI) analysis procedures (Heiri et al., 2001, Dean, 1974). Soil and sediments samples were ground and subsequently homogenized by either stirring by hand or shaking in a closed container. Immediately afterwards, the pow dered samples were placed in pre-weighed crucibles and dried in an oven at 105 C for 24 hours. The dried samples were allowed to cool to room temperature in desiccators a nd afterwards were weighed in order to determine the initial dry wei ght (DW 105) of each sample. Once the initial dry weight was obtained, the samples were heated in a furnace to 500-550 C for 4 hours. Samples were then cooled to room temperature again in desiccators and re-weighed in order to obtain the dry weight of sample (DW 550) in grams. LOI was calculated as follows: LOI550 = ((DW 105 DW 550) / DW 105) 100 Where: LOI550=Loss on Ignition (as percent at 550 C) DW105= dry weight of sample after 24 hours at 105 C DW550= dry weight of sample after 4 hours at 550 C The weight loss determined from this proce dure is proportional to th e amount of organic carbon contained in the sample (Heiri et al., 2001) (see Appendix C for details).


46 Inductively coupled plasma-optical emission spectrometer (ICP-OES) Inductively coupled plasma-optical em ission spectrometer analysis allowed for the determination of Fe and Mn profiles and the likelihood of oxic, suboxic or anoxic conditions at the study sites. The procedure u tilized during this investigation was adapted from EPA Method 3050b (see Appendix D for de tails). The procedure is not a full digestion. It is a strong acid digestion that dissolves almost all elements that could become environmentally available. The ICP-OES method utilizes 1-2 gram s of the homogenized sample. The initial acid addition consists of 10 mL 1:1 HNO3 (1 part DI water to 1 part conc. HNO3). After the acid addition was introduced to the sample and stirred utilizing a glass rod, the mixture was covered with a watch glass. The sample was subsequently heated on a hotplate at 95oC 5o for 10 to 15 minutes, and afterwards allowed to cool. After reaching room temperature, 5 mL concentrated HNO3 was added to the solution. The sample was stirred again, the watch glass replaced and the sample reheated to 95oC 5o for 30 minutes. If brown fumes (an indi cation of oxidation of the sample by HNO3) were generated during this process, the previous step was repeated and another 5 mL of concentrated HNO3 added. Afterwards the sample was returned to the hotplate and the supernatant was allowed to evaporate to approximately 5 mL by heating at 95oC, without boiling, for two hours. Complete evaporation or drying of the samp le was avoided. The sample was allowed to cool to room temperature. Afterward, 2 mL of water and 3 mL of 30 % H2O2 were added. The beaker was covered with a watch glass and warmed to 60oC. Aliquots of 1mL of 30 % H2O2 were added until bubbling subsided Less than 10 mL of 30 % H2O2 were added. The sample was then covered with a watch


47 glass and heated (max of 95oC) until the volume was reduced to approximately 5 mL. Next, 10 mL concentrated HCl was added to the sample, covered and stirred and heated for 15 minutes and then allowed to cool to room temperature. After attaining room temperature, the sa mple mixture was diluted to 100 mL with deionized water and subsequently cent rifuged at 2,000-3,000 rpm for 10 minutes to remove digested particulates. Once centrif uged, another 10 mL of concentrated HCl was added to the sample, which was then heated and stirred for 15 minutes. This mixture was then filtered through a pre-weighed and pr e-labeled Whatman No. 41 filter paper. An aliquot of the filtrate, collected in a 100 mL volumetric flask, was analyzed via the ICPOES instrument. Mineralogical analysis by X-ray diffraction The mineralogical composition of soil a nd sediment samples from the various ecosystems in Puerto Rico was determined using an X-ray diffraction technique (see appendix E). The mineralogical composition of the samples was examined utilizing a Bruker XRD D4 Endeavor. The operating cond itions for the instrument are 20.5 C and 65 % humidity. The step-size and count time used for analyzing th e clay size fraction (<2 m) were 0.02 2Theta and 30 seconds. Sample preparation prior to analysis was performed following and adaptation from Jackson (1958). Samples were treated with an acidic acetat e solution for 24 hours to remove organic matter that may have b een present. Afterwards the sample was centrifuged for 15 minutes at 1300 rpm, and subse quently placed in a water bath at 50 C for 3-4 hours. After being allowed to cool to room temperature, the sample was


48 centrifuged again at 1300 rpm for 15 minutes and treated with hydr ogen peroxide and a NaOAc buffer solution at pH 5 to remove calcium carbonate. This procedure was performed in a water bath for 30 minutes with intermittent stirring. After 30 minutes, the sample was centrifuged and rinsed with 1 N NaOAc. Afterwards a smear of the sample was placed on glass slides and allowed to ai r dry in preparation for XRD analysis. Each clay mineral has a specific feature that can be used for identification via XRD analysis. Peak analysis was initially pe rformed after each sample was air dried and once again 24 hours after the application of ethylene glycol. Peak identification of expandable clays was validated by peak shifti ng which resulted from the ethylene glycol treatment. Illite is a non-expanding group, wh ich is characterized by intense 10-angstrom 001 and a 3.3-angstrom 003 peaks that remain unaltered by ethylene glycol (Poppe et al. 2001). Kaolinite is characterized by a peak at 7 angstroms which will not shift as a result of the glycol treatment (Poppe et al. 2001). Smectite yields XRD patterns that are characterized by basal reflections that vary with humidity. When saturated with ethylene glycol, the 001 reflection of most smectites will swell to about 17 angstroms (Poppe et al. 2001).


49 RESULTS Gamma spectroscopy Downcore Cs-137 activities at Kiani Lagoon and Mosquito Bay range from 0 (or below detection limit) to 2 dpm/g and from 0 to 12 dpm/g, respectively (Figures 9 and 10). Cs-137 activities, in ge neral, are higher at Mosquito Bay, where a maximum activity of 12 dpm/g is located at 1 cm depth. This is likely a result of higher concentrations of clay-size material at the Mosquito Bay si te which led to increased adsorption and retention of radionuclides compared to the Kiani Lagoon site. Cs-137 peaks for Mosquito Bay are located at 1 cm, 20 cm and 25 cm depth. Kiani Lagoon exhibits lower Cs-137 activity with a maximum at 1 cm. It should also be noted that excess Pb -210 data for Mosquito Bay shows a trend of peaks that coincide with Cs-137 peaks thr oughout the core (Figure 10). In terms of activity for excess Pb-210, the activity does not seems to be as high as Cs-137 at Mosquito Bay but it does show peaks that reach 4 dpm/g at that site. Kiani Lagoon excess Pb-210 activity reaches a maximum valu e of 9 dpm/g. The higher activity is located at the surface of the co re. The activity in general decreases with depth. There were three sub surface peaks found in this co re at 18 cm, 24 cm and 34 cm depth (Figure 9). Inventory calculations for Kiani and Mosquito Bay demonstrate that Kiani Lagoon has a higher Cs-137 inventory (0.1 dpm/cm2) when compared with Mosquito Bay


50 (0.04 dpm/cm2). If it is assumed that the initial distribution of the Cs-137 fallout input was uniform, the deviations in the measured distribution of Cs-137 from the local fallout inventory should represent the net impact of soil redistribution during the period following Cs-137 deposition. Lowest Cs-137 activities were determined for surface sediment and soil samples collected from Green Bay (below detection limit) and Baracuda Lagoon (below detection limit), respectively (Figure 11). Cs-137 activit ies in surface samples collected at these two locations are generally below detection limits. The highest su rface Cs-137 activity, 0.1 dpm/g, was determined in a soil sample collected from the Operational Posting Lagoon ( Fig.11). Cs-137 activities for surface samples collected from mangrove forest environments ranged from 0.036 dpm/g at S. Jobalos to 0.0198 dpm/g at Kiani Lagoon (Table 8, Fig.11). Figure 8. Downcore Cs-137 and exPb-210 Kiani Lagoon, Vieques, PR. 0 5 10 15 20 25 30 35 0246810Depth (cm)Activity (dpm/g) Cs-137 exPb-210


51 Figure 9. Downcore Cs-137 and exPb-210 Mosquito Bay, Vieques, PR. Figure 10. Surface samples gamma analysis and clay content 0 5 10 15 20 25 30 024681012Depth (cm)Activity (dpm/g) Cs-137 exPb-210 0 0.5 1 0 0.02 0.04 0.06 0.08 0.1 0.12 02468Activity Cs-137 (dpm/g) Sediment Volume (%)Surface Sample Vieques Clay Cs-137


52 Downcore profiles for Cs-137 at the BO NUS facility area for Station 1 and Station 2 show activities that range from 4 dpm/g to 12 dpm/g and from below detection limits to 1 dpm/g respectively (Figure 12 and 13). Station 1 profile shows two peaks, the first one at 6-8 cm depth (7 dpm/g) and th e second peak at 10 cm depth (12 dpm/g) (Figure 12). The exPb-210 pr ofiles show an activ ity that decreases as a function of depth. One peak stands out at 10 cm depth, wh ich also coincides with the highest activity peak of Cs-137. Figure11. Downcore Cs-137and exPb -210 at BONUS St. 1, Rincn, PR 2 4 6 8 10 12 14 16 18 051015Depth (cm)Activity dpm/g Cs-137 exPb-210


53 Figure 12. Downcore Cs-137and ex Pb-210 at BONUS St.2, Rincn, PR Cores samples from Station 2 of the BONU S facility exhibit Cs -137 activities that are less than those determined at Station 1. In this core, most sample increments displayed activities that ha ve zero activity of Cs-137 a nd four increments showed activities that ranged from 0.2 dpm/g to 0.4 dpm/ g. Cs-137 activities determined at all of the other stations at the BONUS area were higher than those observed at Station 2 (Figure 13). At Station 1, th e highest activity is 12 dpm/g ( 10 cm depth). In contrast, at Station 2 Cs-137 activity does not exceed 0.5 dpm/g. 0 1 2 3 4 5 6 (cm)Activity (dpm/g) Cs-137 exPb-210


54 Figure 13. Downcore Cs-137and exPb -210 at BONUS St. 3, Rincn, PR Figure 14. Downcore Cs-137and exPb -210 at BONUS St. 4, Rincn, PR Maximum activities for cores at stations 3 and 4 are in 9 dpm/g at 4 cm depth and 5 dpm/g at 1 cm depth (figures 14 and 15). On e peak of Cs-137 (5 dpm/g) was found at 4 cm of the core at station 3, activity throughout the re st of the core fluctuated between 0.5 0 5 10 15 20 25 051015202530Depth (cm)Activity (dpm/g) Cs-137 exPb-210 0 1 2 3 4 5 6 0123456Depth (cm)Activity (dpm/g) Cs-137 exPb-210


55 dpm/g to 5 dpm/g ExPb-210 at station 3 e xhibited higher activity peaks (a maximum of 27 dpm/g) in the upper few centimeters of the core. Only two increments displayed Cs137 and exPb-210 activities that were belo w detection limits. Station 4 was the shallowest station at the BO NUS site. However, Cs-137 act ivity levels at the site exhibited two peaks for Cs-137, the first one at 1 cm (5 dpm/g) and another at 5 cm depth (5 dpm/g) (Figure 15). The exPb-210 showed two peaks, one of 5.5 dpm/g at 0.5 cm and another at 2 cm of 4 dpm/g. Cs-137 inventorie s calculated for all stations at the BONUS facility ranged from 1.2 dpm/cm2 to 11 dpm/cm2. Inventories for stations 1-4 were as follows, 11, 1.2, 6.7 and 5.4 dpm/cm2, respectively. Figure 15. Downcore Cs-137 and exPb-210 at El Verde St. 1 Rio Grande, PR. 0 5 10 15 20 25 02468101214Depth (cm)Activity (dpm/g) Cs-137 exPb-210


56 Figure 16. Downcore Cs-137and exPb-210 at El Verde St. 2 Rio Grande, PR. Downcore gamma spectroscopy analysis was performed on cores collected from the CNF’s El Verde Experimental Station and Espiritu Santo Estuar y (Figures 16-19). Results differed from site to site. Sample s obtained from the El Verde Experimental Station were collected from various elevations, ranging from 250 to 500 m above sea level. Mean annual rainfall for this region is 3456 mm, and monthl y temperatures range from 21-25 C. The Espiritu Santo Estuary samples were collected from the estuary; station 1 at an open beach area and station 2 in a low impact environment. Figure 17 shows Cs-137 activity was higher in the uppe r 5 cm of the core at the El Verde Experimental Station 2. At st ation 2 the higher activities we re 3 dpm/g (9 cm below the surface) and 4 dpm (11 cm below the surface). At station 2, higher activity was located at the top of the core, 12 dpm/g (at the surf ace), 8 dpm/g (2 cm below the surface) and 10 dpm/g (4 cm below the surface). There we re three and eight dow ncore segments in 0 2 4 6 8 10 12 14 051015Depth (cm)Activity (dpm/g) Cs-137 exPb-210


57 which Cs-137 activities were below detec tion limits at station 1 and station 2, respectively. ExPb-210 activity was generally higher than Cs-137 activity in bot h cores. But at El Verde station 2 exPb-210 activ ities decreased with depth. The trend was not visible at El Verde station 1, where excess Pb-210 ac tivity ranged from 4 dpm/g to 12 dpm/g. There were 2 increments where exPb-210 act ivity was below detection limit. At El Verde station 2, samples ranged from 8 dpm/g (at the surface) to 1 dpm/g (at the bottom of the core). Figure 17. Downcore Cs-137 and exPb-210 at Esp. Santo Estuary St.1 Rio Grande, P.R. 0 10 20 30 40 50 60 05101520Depth (cm)Activity (dpm/g) Cs-137 exPb-210


58 Figure 18. Downcore Cs-137 and exPb-210 at Espiritu Santo Estuar y St.2 Rio Grande, P.R The pattern observed at El Verde stations 1 and 2 was not observed at the Espiritu Santo Estuary (ESE). Station 2 was located on an open beach. The core collected at ESE station 2 possessed many increments in which Cs-137was below detect ion limits. In fact, Cs-137 activity was below det ection limits for 12 out of 23 sample increments. ESE station 1 is located within a red mangrove prot ected area, near the mouth of the river. The Cs-137 activity for this core was above detection limits for all sample increments. The average Cs-137 activity level at ESE wa s 8 dpm/g for station 1 and 0.02 dpm/g for station 2. The highest activity for Station 2 was 2 dpm /g (at the surface of the core). The highest activity for station 1 was 15 dpm/g ( 19 cm below the surface). Cs-137 activity is present at ESE station 2, a robust peak is s hown at the upper centimeters of the core and 0 2 4 6 8 10 12 14 16 18 00.511.522.5Depth (cm)Activity (dpm/g) Cs-137 exPb-210


59 the remaining of the core only shows how the activity decreases as a function of depth. Several Cs-137 peaks were observed at stati on 1. It should be noted that the core obtained for station 1 was longer than th at which was obtained for station 2. Data from these four rainfo rest stations revealed an elevation gradient. This gradient could be an important factor when an alyzing the total activity of the cores. For instance, surface samples collected from th e radiation center and the injection area showed the highest activity of 22 dpm/g. There are some surface samples that also showed Cs-137 activities that were below de tection limits. On average, the highest activities were found at the highest elevatio ns of the rainforest study sites. One interesting fact is that alt hough the activities increased with higher elevation, the higher inventory was located at the Espiritu Santo Estuary station 1, with an inventory of 22.92 dpm/cm2 (figure 18). At higher elevations of El Verde, station 1 had an inventory of 10 dpm/cm2 and station 23 dpm/cm2 (figure 19). Values for ESE station 1 versus the inventory value for ESE station 2 represent a substantially larger c oncentration of Cs-137 activity (figure 19). Inventories for the sta tions located at the hi gher elevation did not show such a difference in the inventory. Stat ion 1 of El Verde is 3 times higher than the Cs-137 activity at station 2 which is located at roughly the same elevation (figure 19).


60 Figure 19. Cs-137 Inventories fo r Stations at the El Verde and Espiritu Santo Estuary. Figure 20. Correlation between Cs-137 activity on surface samples from the radiation center at El Verde Experimental St ation and finer partic le size distribution 0 5 10 15 20 25 EV St. 1EV St. 2ESE St. 1ESE St. 2Cs-137 Inventory (dpm/m^2) 0 5 10 15 20 25 0 10 20 30 40 50 60 70 80 90 051015202530Activity Cs-137 (Bq/g) Sediment Vol (%)Sample Number Clay Silt Cs-137


61 Grain size analysis Vieques surface samples were collected from a variety of ecosystems located within restricted areas of the Vi eques National Wildlife Refuge (VNWR). Fine grain clay-rich sediments possess a greater ability to absorb and retain radionuclides than sediments dominated by la rger particles. Grain size analysis indicates that the average concentration of clay size particles varied at each location and the average concentrations at Kiani Lagoon (9 % clay) was slightly higher than at Mosquito Bay (6 % cl ay) (Figures 22-23). Maximum silt size concentrations were determined in th e upper 5 cm of Kiani Lagoon Station 1 (65 %). A core collected from Mosquito Bay contained maximum silt size particles distribution of 35-40 % (s tation 1) in the upper 5 cm of sediments (Figure 23). Average sand size concentrations were hi gher at the Mosquito Bay (73 %) than at Kiani Lagoon (63 %). Figure 21. Grain size di stribution at Kiani Lagoon Station 1, Vieques, PR 0 5 10 15 20 25 30 35 020406080100Depth (cm)Grain Size Distribution (%) Silt Clay Sand


62 Figure 22. Grain size distri bution at Mosquito Bay La goon Station 1, Vieques, PR BONUS Area, Rincon, grain size characteristic s were fairly consistent for all of the BONUS stations. The clay distribution is the smallest component, followed by silt and then sand, which was the most preval ent particle size at the BONUS sample locations. Clay distributions ranged from le ss than 10 % at Station 1, to almost 30 % at Station 4 (Figures 24-27). Silt distributions range from 20 % to 50 % for the sampled areas. Silt size distributions were higher than the sand component at Station 4. 0 5 10 15 20 25 30 020406080100Depth (cm)Grain Size Distribution (%) Silt Clay Sand


63 Figure 23.Grain size distribution at BONUS Area, Station 1, Rincn, PR Figure 24. Grain size distribution at BONUS Area, Station 2, Rincn, PR 2 4 6 8 10 12 14 16 18 020406080100Depth (cm)Grain Size Distribution (%) Silt Clay Sand 0 2 4 6 8 10 12 14 020406080Depth (cm)Grain Size Distribution (%) Silt Clay Sand


64 Figure 25. Grain size distribution at BONUS Area, Station 3, Rincn, PR Figure 26. Grain size distributions at BONUS Area, Station 4, Rincn, PR 0 5 10 15 20 25 020406080Depth (cm)Grain Size Distribution (%) Silt Clay Sand 0 1 2 3 4 5 6 0102030405060Depth (cm)Grain Size Distribution (%) Silt Clay Sand


65 At the Caribbean National Forest ESE station 2, clay size particles had a fairly constant distribution that averaged 1.75 % (figure 29). The highest clay size particle distribution value at station 2 was 4 % at a depth of 6 cm and the lowest value was 0.75 % (determined at a depth of 3 cm). Silts range between 10 % to 33 % and its average distribution was 13 %. Sand was the most common particle size in the core. Its distribution range from 63 % to 88 %. The hi ghest percentage of sand was found at 2 cm depth and the lowest value at 6 cm depth. Data for ESE station 1 were very similar for the clay distribution, but silt was the most abunda nt particle in this core. Average clay distribution was higher at stati on 1 (figure 28) with 7.6 % clay distribution. Silt’s highest value was 88 % at the top of the core of station 1 and the lowest value was 59 % at 33 cm depth. Sand distribution was the lowest of t hose found at this locati on with an average of 9 %. Sand size particle distri bution ranged from 2 % at 2 cm depth to 36 % in the middle part of the core. Figure 27. Grain size distribution at Espiri tu Santo Estuary St. 1, Rio Grande, PR 0 5 10 15 20 25 30 35 40 45 50 55 60 020406080100Depth (cm)Grain Size Distribution (%) Clay Sand Silt


66 Figure 28. Grain size distri bution at Espiritu Santo Es tuary St. 2, Rio Grande, PR Figure 29. Grain size di stribution at El Verde St. 1, Rio Grande, PR 0 2 4 6 8 10 12 14 16 18 20 020406080100Depth (cm)Grain Size Distribution (%) Silt Clay Sand 0 5 10 15 20 25 020406080Depth (cm)Grain Size Distribution (%) Silt Clay Sand


67 Figure 30 displays grain size data for El Verde station 1. The size distributions for stations at higher altitude were not as constant as the estuarine stations. Silt distribution in these soil samples seemed to decrease with depth and the sand distribution increased. Clay distributi on at this site ranged from 7 % to 25 %. The uneven distribution among increments was more obvi ous on silt and sand particles since sand distribution went from 6 % at 1.5 cm to 73 % at 19 cm depth. These values showed a slight increase with depth. However sand di stribution did not constantly increase as it showed increments where its percentage went as low as 15%. In the case of silt, its distribution had some fluctu ation throughout the core but a tendency to decrease as a function of depth was demonstrated. Figure 30. Grain size di stribution at El Verde St. 2, Rio Grande, PR 0 2 4 6 8 10 12 14 020406080100Depth (cm)Grain Size Distribution (%) Silt Clay Sand


68 Station 2 at El Verde (fi gure 31) did not possess the sa me pattern as station 1. Station 2 had an average clay size distributi on of 16 % which was close to the distribution calculated for station 1. There was no decrease with depth at station 2, as was shown at station 1. Clay and sand seemed to follow th e same decreasing trend. Average sand and silt values were 44 % and 23 % respectively. Silt ranged from 16 % to 62 %. These two values, half a centimeter apart, represent th e highest and lowest values of the core. Seeing these values in such a close incremen ts confirmed the high level of inconstancy within the soil located in the rainforest ar ea. Sand size particle distribution ranged from 17 % to 79 %. Again these increments were just half a centimeter apart from each other. Loss on ignition (LOI) LOI is commonly used to estimate the percent of organic matter present in a sample (Henri, 1970). Organic matter conten t was analyzed to determine how Cs-137 behaved in its presence. St udies have shown that Cs-137 could be retained by organic matter (Santchi and Honeyman, 1989). The highest LOI values (e.g. organic matter concentration) were determined for Kian i Lagoon samples. A maximum of 35 % LOI was determined within the upper 5 cm of this co re (Figure 32). LOI results indicated that the organic matter content at Kiani Lagoon and Mosquito Bay decreased as a function of depth. Maximum LOI values determined fo r Mosquito Bay were above 10 %. The highest organic matter content was found with in the upper 5 cm of both cores (Figure 32).


69 Figure 31. Organic matter percentage at Ki ani Lagoon and Mosquito Bay, Vieques, PR. The organic matter content (represente d by LOI) for the BONUS facility area, figure 33 indicated a heterogeneity among samp les at the study site. Organic matter content ranged from 2 % to 13 %. Station 2 presented the lowest organic matter content (ranged from 6 % at the top of the core to 3 % at the bottom). The highest organic matter content was presented at stat ion 4, where it ranged from 11% and 13%. Organic matter content at the other two stations ranged from 2 to 10 %. The same stations that possessed higher distributions of fine grain material also possessed higher percentages of organic matter (figure 33). 0 5 10 15 20 25 30 35 40 010203040Depth (cm)Organic Matter (%) Kiani Lagoon … Mosquito Bay …


70 Figure 32. Organic matter percenta ge at BONUS Area, Rincn, PR. Figure 34 shows results for analyzed sample s at 4 stations in the tropical rain forest area and the ESE. As seen in th e figure 34 the organic matter concentration ranged from less than 5 % to 35 % organi c matter. Samples collected from sandy environments possessed the lowest values for organic matter and those that were collected from protected environments possessed intermediate organic matter concentrations. Samples collected from soils at the radiation center and the injection area possessed the highest LOI values. LOI values for surface samples from these latter sites ranged from 11 % to 61 % (figure 34). Station 14 (figure 34), at the radiation center, had the highest organic content of all the samples collected from the rainforest and station 5 had the lowest organic content of all samples. 0 2 4 6 8 10 12 14 16 18 20 02468101214Depth (cm)Organic Matter (%) R. St.1 R. St. 2 R. St. 3 R. St. 4


71 Figure 33. Organic matter in percentage at El Verde and Espiritu Santo Estuary, Rio Grande, PR. Figure 34. Organic matter in percenta ge at El Verde Surface Samples 0 10 20 30 40 50 60 010203040Depth (cm)Organic Matter (%) ESS St. 1 ESS St. 2 EV St. 2 0 10 20 30 40 50 60 70 57911131517192123252729 StationNumberat theRadiationCenter Organic Matter ( %)


72 Inductively Coupled Plasma-Optical Emission Spectroscopy (ICP-OES) Results from ICP-OES provided informati on regarding the con centration of more than a dozen of elements that are useful to determine the oxic, suboxic or anoxic conditions of the sediments (figure 35). Sa mples from Mosquito Bay indicated a steady increase of Fe and Mn concentration with depth. The highest iron concentration was located at 18 cm depth, 9356 ppm. The lowe st iron concentration was present at 4 cm depth, 960 ppm. Manganese concentration wa s consistently low throughout the length of the core when compared with iron concentra tions. However, Mn concentration increased as a function of depth until the 17 cm depth where its concentration rose to 38 ppm. Below this depth its concentration decreased with depth throughout the remainder of core. In addition, there were also 5 sediment increments for which the Mn concentrations were below detection limits. Due to oxic a nd suboxic states within the sediments, iron concentrations ranged from 2753 ppm (at th e top of the core) and 8463 ppm (at the bottom of the core) (figure 36). The Mn values ranged from 1 ppm at the top of the core to 18 ppm at the bottom of the core. Samples analyzed from Kiani lagoon showed a constantly low manganese concentration throughout the depth of the pr ofile and its values ranged from below detection limits to 70 ppm. Iron concentrati on seemed to decrease with depth with a slight elevation in concentra tion located near the bottom of the core (a maximum of 6086 ppm). At 3 cm and 19 cm, Fe concentr ations were 25072 ppm and 20159, respectively which could be attributed to the oxic and suboxic states mentioned above (figure 36). Surface samples from Vieques (figure 38) pr esented the same trend in Fe and Mn concentration. Mn possessed th e lowest concentration at al l samples and Fe was present


73 in higher concentration. The OP entrance st ation was the area where Fe had a highest concentration, 16, 033 ppm. Figure 35 Downcore concentration for Ir on and Manganese at Mosquito Bay St. 1, Vieques. Figure 36. Downcore concentration for Ir on and Manganese at Kiani Lagoon St. 1, Vieques. 01020304050 0 5 10 15 20 25 050001000015000 Concentration Mn (ppm)Depth (cm)Concentration Fe (ppm) Fe 020406080100 0 5 10 15 20 25 0100002000030000 Concentration Mn (ppm)Depth (cm)Concentration Fe (ppm) Fe


74 Figure 37. Iron and Manganese concentrati on for surface samples at Vieques National Fish and Wildlife Refuge, Vieques, PR. ICP-OES analyzed samples are s hown in figures 39-40, manganese concentrations throughout the core at all stations ranged from 0-596 ppm. The highest concentration was present at station 4 at a depth of 1.5 cm. Manganese concentration in soil samples from the BONUS area ranged from 150 ppm to almost 600 ppm. Iron concentrations at BONUS ra nge from around 8000 ppm to 16,000 ppm. Station 3 possessed the highest iron concentration. Higher Fe concentrations were located in the upper centimeters of the core. The highest value was 13,783 ppm at 5 cm. The average iron concentration throughout the cores was13,000 ppm. At station 3 and 4 the difference in concentrations was similar to those discussed in the previous core, however, these profiles showed simi lar behavior (figures 39 and 40).


75 Figure 38. Downcore concentration for Iron and Manganese at BONUS Area St. 3, Rincn, PR. Figure 39. Downcore concentration for Iron and Manganese at BONUS Area St. 4, Rincn, PR. 150250350450550 0 2 4 6 8 10 12 14 16 18 20 800010000120001400016000 Concentration Mn (ppm)Depth (cm)Concentration Fe (ppm) Fe M … 0200400600800 0 1 2 3 4 5 6 05000100001500020000 Concentration Mn (ppm)Depth (cm)Concentration Fe (ppm) Fe Mn


76 Manganese and iron concentrations from the Caribbean National Forest are presented in figures 41 through 43. Manganese concentrations in soil and sediments remained consistently low. The iron concen tration decreased with depth and ranged from 28,632 ppm to 15,441 ppm. Stations at ESE showed an interesting difference in concentration levels. Fe concentrations at Station 2 were one or der of magnitude smaller (3,266 ppm to 2,866 ppm) than station 1 (35,638 ppm to 21,066 ppm). However, Fe concentrations at both locations decreased with depth. Values for th ese stations were also higher at the top of the core in both instances. Figure 40. Downcore concentration for Ir on and Manganese at El Verde St. 1, Rio Grande, PR 050100150 0 2 4 6 8 10 12 05000100001500020000250003000035000 Concentration Mn (ppm)Depth (cm)Concentration Fe (ppm) Fe M …


77 Figure 41. Downcore concentration for Iron and Manganese at Esp. Santo Estuary St. 1, Rio Grande, PR Figure 42. Downcore concentration for Iron an d Manganese at Espir itu Santo Estuary St. 2, Rio Grande, PR 0100200300400 0 5 10 15 20 25 30 35 40 010000200003000040000 Concentration Mn (ppm)Depth (cm)Concentration Fe (ppm) Fe Mn 01020304050 0 2 4 6 8 10 12 14 16 1000150020002500300035004000 Concentration Mn (ppm)Depth (cm)Concentration Fe (ppm) Fe Mn


78 X-ray diffraction Figure 43. Mineralogical analysis of bulk samples from Vieques, PR X-ray Diffraction (XRD) data for Mosquito Bay indicated the ab sence of illite at this location (figure 44). However, smectite a nd kaolinite were present in the majority of downcore samples collected at the site. Highe st Cs-137 activities at this station were observed in samples in which kaolinite was present and illite and smectite were absent. Surface samples collected at the different locations within the VNFWR (Figure 20) were also analyzed using the XRD t echnique. The XRD analysis confirmed the presence of kaolinite in all samples and the absence of illite at th e study sites (figure 43).


79 Figure 44. Mineralogical anal ysis of bulk samples from the BONUS area, Rincon, PR Mineralogical analysis of the BONUS ar ea was performed on a subset of samples that were identified based upon Cs -137 activities. Samples with the highest ac tivities, as well as samples with the lowest activities were analyzed for the purpose of detecting the absence or presence of illite, kaolinite and smectite. Clay mineralogy results were similar for all stations at the BONUS facility area. Illite was absent throughout the cores, smectite and kaolinite were present (figure 44).


80 Figure 45. Mineralogical analys is of bulk samples from El Verde and Espiritu Santo Estuary X-ray diffraction conducted at the rainforest area as well as the estuarine area is shown in figure 45. Results from the analysis revealed the absence of illite in all sample increments. The presence of smectite and kaolinite was determined for almost all analyzed samples that were collected from this study area. Samples with high Cs-137 activity at ESE Station 1 of the estuarine ar ea possessed smectite, but did not indicate the presence of kaolinite or illite. Samples an alyzed from this stu dy site which possessed low Cs-137 activity values contained kaolinite All samples that were analyzed from station 2 possessed kaolinite and smectite. Illite was completely absent from the soil samples collected from stations 1 and 2. Kaolinite was present at all stations.


81 DISCUSSION The mineralogical composition of the study sites is a reflection of the complex nature of the tropical ecosystem. Both mi neralogical analysis, X -ray diffraction and the grain size distribution, provide d information that Vieques, Rincon and the El Verde Experimental Station soils are composed of clay particles that include smectite, kaolinite, chlorite and vermiculite. Illite was completely absent from the mineralogical composition of the study sites. Soils, in genera l, at each study site possessed a grain size distribution that could be described as constant. Downcore clay size particle distribution was fairly constant with depth at all study sites. It compre sses the lowest percentage of particles sizes at all sites except for a few st ations at the BONUS where clay distribution was as high as 20 %. This may be the resu lt of the undisturbed environment from where the sample was collected. Clay distribution in the soils at the B ONUS facility area was low. Since Cs-137 is generally associated w ith finer grain sediments this may be one of the reasons why Cs-137 does not appear to have been retained. There are only two stations at which clay size particles exceed ed 10 %. Stations 2 and 4 possessed more clays than any of the other stations at the BONUS facili ty. Station 4 possessed Cs-137 activities that were one order of magnitude hi gher than station 2. Therefore, it appeared as though for this particular study site, that so il exhibiting more clay size particles did not necessarily possess higher Cs-137 activities. Station 2 is in a protected envir onment surrounded by mangroves and tidal influence in which the sedimentation process ge nerates a higher density of clays. Station


82 2 from Espiritu Santo Estuary would more lik ely retain Cs-137 coming from the slopes of the rainforest. Indeed this station is th e only station where Cs-137 was found throughout the entire core. However, its clay distribu tion remained between 10-15 %. At the higher station of El Verde, clay dist ribution was higher than the distri bution at the estuarine area. The difference in this fact might be due to the environmental processes such as weathering of rock in the upper elevations, which plays an important role in the formation of mineral particles. Weathering can be influenced by rain received in the area, which play a role in erosi on and fluvial sediment movement. The observed higher Cs-137 activity levels in the surface samples from El Verde were directly related to the clay content of the soil there. There was no trend in the downcore Cs-137 activities or grain size distributions determined at El Verde Experimental stations 1 and 2. The surface samp les, as well as the soil samples, had a higher content of clays but there was no consiste ncy of the composition of the soil layer. The dominant soil form was silt, which could provide binding sites for pollutants, but not to the same extent as clay sized particles. In Vieques, clay size particles did not exceed 10 % for any sediment interval at either sample location. Downcore exPb-210 was slightly higher in the Mosquito Bay core than in the Kiani Lagoon core (figures 910). This was an indi cation of the different environments from which the samples were collected. The Kiani Lagoon is a calm aquatic environment in which the only input of water is the rainfall and some small tidal influence. The Mosquito Bay is an open coastal area in which physical processes contribute to the movement of sediment particles in the water column. Surface samples from Vieques also indicated the same pattern where clay distribution was low and silt and


83 sand size particles dominated the local mineralogy. This is therefore indicative that although there was fewer fine size sediment particles present w ithin the Kiani Lagoon core compared to the Kiani Lagoon core, the fine r particles at the Mosquito Bay site were able to more effectively absorb Cs-137 as th e result of global fallout and later introduced to Vieques’ aquatic environment. It has been shown that humic (colloidal ) materials increase the retention of radionuclides in the ocean (San tschi et al; 2006). The presen ce of higher organic matter content (as estimated by LOI) may have incr eased the chance for radionuclide retention capabilities of the local sediments resulting in higher Cs-137 activit ies. Organic matter content was different at th e different ecosystems where the study was conducted. For instance in the higher altitude of the rainforest, more dry leav es and dead trees highly increased the organic matter content of soils. However, Espiritu Santo River station 1 is located in a mangrove protected environment which promoted less input of debris and other sources that increased th e organic matter. Station 2 at the Espiritu Santo River is located in an open beach environment and th erefore the organic matter content there did not exceed 5 %. In a nearby protected mangrove forest, coastal influence on the accumulation of organic matter was noticeable at station 1 which organic matter content exceeded 20 %. The estimation of the organic matter conten t obtained by LOI at areas such as the BONUS site also confirms that organic matte r content will vary based on the ecosystem where the sample is collected. LOI samples fr om the BONUS areas remained fairly low. The lowest organic content was at station 2 which was closer to the shore. There the high energy waves disturbed the sediments as well as the accumulated soils. High waves


84 also formed during cold fronts and the subs trate was washed away bringing sand from beach shore to the area where samples were collected. Organic content in samples from Vieques i ndicated that, at the surface of the core, higher organic content might be the result of the debris and buried shells from dead organism. Therefore Kiani has a higher organic content when compare with the Mosquito Bay area. Surface samples from Vieques had similar downcore mineralogy, but the samples collected from surface areas had higher organic content. Atmospheric deposition of Cs-137 as a re sult of global fallout began in the 1950’s. Cs-137 was present in the deepest sedime nt intervals at most sample locations, indicating that these sediments were deposited after the onset of atmospheric fallout. As shown by gamma spectroscopy data (Figure 11-14) high to low Cs-137 activity trends were found in soils adjacent to the BONUS facility. This overall trend is likely the result of the characte ristically slow mobility of Cs-137 on a substrate (Forsberg et al; 2000). Cs-137 measured in the soil collected adjacent to the BONUS facility maybe the result of global fallout and not th e result of the flooding and leakage from the remains of the power plant. Although floodi ng occurred at the plant during hurricane George, the activity levels measured in the upper centimeters of the soil did not exceed the values measured by Alonso-Hernandez (2002, 2006) in Cienfuegos, Cuba or the fallout rate measured in Port St. Luci e, Florida (DHS, 2008). If the 1998 flooding disturbed the entombment and released a porti on of the remaining Cs-137 to the adjacent area, it would be expected that higher Cs137 activities would be present in the local soils, especially near the surface. The natural downhill slope upon which the power plant


85 was built should also be taken into consideration. This slope would have served to facilitate the removal of the contaminants from the surface soil through runoff events. If runoff occurred and radionuclides were tr ansported within the sediments, it is likely that radionuclides would be transported further into the water column. It has been shown that sand does not absorb radionuclides due to its coarse grain and the high permeability of the minerals. The high energy beach adjacent to BONUS is mainly composed of sand and coarse grain material wherein radionuclides would percolate and their retention will be minimal. In the ev ent that a leakage did occur and Cs-137 was released from the entombment to the local environment during tr opical storms, it would be hard to retain Cs-137 in the soil due to the high content of sand and low clay distribution. There are three intervals between 5 and 25 cm in which the level of Cs-137 present in the core was below detection limits for the Kiani Lagoon stations. Eight intervals within the upper 15 cm of sediment of the Mosquito Bay core exhibited either the absence of Cs-137 or the presence of Cs -137 at levels that were below detection limits. Intermittent intervals of sediment in which Cs-137 was either absent or below detection limits was likely the result of: 1) biogeochemical processes, such as Cs-137 desorption caused by cation competition result ing from anoxic conditi ons or salt water intrusion, or 2) the absence of permanent Cs -137 particle binding si tes (e.g. interlayer position of illite). Although sediment samples that were analyz ed did not contain illite, the presence of smectite and kaolinite was detected. The su rfaces of kaolinite and smectite, as well as the interlayer position of smectite potentially served as non-specific binding sites for Cs-


86 137. When compared to other clay minerals, smectite has the greatest ability to absorb Cs-137, as well as other cations, as a result of its expandable interlayer position. However given the fact that the absorption ab ility of smectite is non-selective, in the presence of increased dissolved cation con centrations (e.g. higher salinities or anoxic conditions) Cs-137 that was previously bound to smectite particles could be easily desorbed from the interlayer position and re introduced in the dissolved phase into the local aquatic environment. Since radionuc lides in Vieques are present in oxic environments, their mobility will depend on sediments and water transport. As we can see on Figure6-9, there is no evident trend on th e distribution of Cs-137 in Vieques. This can be the result of sediment disturbance and mixing which was affected by the process of sedimentation. In the eventuality of a hurricane, sediment mixing will occur, making it harder to obtain a geochronology that represents the reality of the island of Vieques in terms of Cs-137 and other contaminant retention in sediments. Figure 46. Cs-137 fallout record for the Miami region, adapted from Surface Air Sampling Program, Department of Homeland Security, 2008. 0 1000 2000 3000 4000 5000 6000 1955196019651970197519801985199019952000 YearCs-137 Activity (uB/sq m)

PAGE 100

87 Cs-137 activities determined during this investigation did not exceed Cs-137 activities determined in the state of Florid a. Fallout Cs-137 at Ft. St. Lucie is 29.6 Bq/m3. Fallout data for Miami is shown in figure 47; the oscill ations that were observed on the sampling sites in Miami did not match the data presented here in the current project. Environmental conditions and exposure to tidal and weather influences definitely influence the retention capacity of the ecosy stem. The inventories of station 2 are 20 times higher than the inventory from the b each ecosystem where current and deposition influence clay and organic matter accumulation. The study sites are composed of oxic substr ates where Cs-137 mobility is slow. Factors affecting the retenti on of radionuclides are the grai n size and organic content of the substrate as well as the impact received in the environment. Due to the effectiveness of the binding site and the high cation competition for binding sites, Cs-137 is present at low concentrations at the estuary and the ra inforest ecosystem of the island of Puerto Rico. Some other factors controlli ng the fate and transport of Cs-137 at all study sites of this project might be the variable mixing at Vieques and the Espiritu Santo Estuary. As seen on Figures 6-9 and 15-18, gamma spectro scopy profiles confirme d the presence of Cs-137. Variable mixing enabled the sedime nts to be deposited in a non-uniform way, thereby providing the stratigra phic pattern we observed. Bioturbation can also affect sediment distribution. As marine organisms live in the sediment column, they scavenge the se diments looking for their food source. As

PAGE 101

88 benthic fauna make their way to the top or the bottom of the sediment column, radionuclides can be moved from the depos itional system. In the current study movement could have been caused by benthic organisms as they removed organic matter and clay minerals that provide d the binding site for Cs-137. Sediments coming from the rainforest area have been heavily disturbed by the development and the wide range of sedimenta tion rates. The Espiri tu Santo Estuary has been a major location for local housing and tourist development. During the last 10 years, more than 300 hundred houses were built in the area as well as a new resort hotel complex which includes a golf cour se that is adjacent to the Espiritu Santo River. Urban development will increase run-off into the estuarine area which will contribute to the faster accumulation of sediments in th e area. Webb and Gomez (1998) provided information on sedimentation rates for a si milar area and the rates ranged from 0.24 to 3.9 cm per year of sediment. Sedimentation ra tes will definitely be higher at the Espiritu Santo Estuary where all the water from runoff comes to the bay. Landslides are also very common in the rainforest and as they o ccur, more sediment will quickly be added to the water column. As such, if Cs-137 were released to the so il, it would likely be bound to smectite or kaolinite, two clay minera ls that unlike illite bind Cs-137. Given the absence of 10 Angstrom interl ayer spacing of illite that repr esents strong i rreversible Cs137 binding sites, smectite which possesses an interlayer space of 10-18 Angstrom would likely serve as the major binding material of Cs-137 in the local environment. The presence of other elements that can out compete Cs-137 for binding site would depend on the interstitial space. Smectite, which has a larger interlayer space than illite, would facilitate the binding of other elements such as K which will then outcompeting Cs-137.

PAGE 102

89 Cs-137 absorption on kaolinite is limited to the su rface of this clay mineral. Like that of smectite, Cs-137 binding onto kaolinite is non-selective and can be temporary. The ICP-OES results indicate that soils at the BONUS facility are present in a primarily oxic environment. Studies have s hown that anoxic environments facilitate the mobilization of radionuclides in the substr ate. When Cs-137 is present in oxic environments, its mobility will be slower co mpared to the anoxic environment. The presence of organic matter is higher at st ation 5 and the same station 5 showed the highest Cs-137 activity. Taking in to consideration that illite is not present at any of the stations at BONUS and the substr ate is oxic, it will be logical to infer that organic matter is playing an important role in the rete ntion of Cs-137 at the studied sites. This investigation provides insight into the radiogeochemistry of the rainforest and the adjacent Espiritu Santo Estuary. Ga mma spectroscopy revealed the presence of Cs-137 at all sites. A difference in the Cs137 activity levels determ ined at the various locations can, in part, be attributed to the different types of environments.

PAGE 103

90 CONCLUSIONS Cs-137 activities in Vieques sediments and soils vary with location. The distribution of Cs-137 in surface samples colle cted from mangrove forests, marsh and lagoon environments on the island range from below detection limits to 0.6 dpm/g. There is no correlation between type of e nvironment (e.g. mangrove forest, marsh, and lagoon) and Cs-137 activity present in Vieque s surface sediments and soils. Downcore data suggest that there are fine grain size particles pres ent at both Mosquito Bay and Kiani Lagoon which are able to effectively absorb and retain radionuclides introduced into these local environments. Downcore Cs-137 activity profiles indicate the occurrence of previous sporadic epis odic events (e.g. mixing, burial, erosion, bioturbation, hurricane/storm/wind precipitate d particle removal, adsorption and absence of suitable particle binding sites) have tr ansported Cs-137 laden particles away from original deposition sites. The absence of illite may have influenced the retention capacity of Vieques sediments. The retention of Cs-137 might be influenced by the organic matter present in the analyzed samples. The presence of the clay mineral particles kaolinite and smectite, which provide a weaker binding site for Cs-137 to the sediments, was observed. Sediments in Vieques are oxic sediments which also influence the slow movement of Cs137. Other physical and biological paramete rs like sediment disturbance due to hurricanes and bioturbation can affect sediment mixing.

PAGE 104

91 Findings from this study provide insi ght regarding the extent to which radionuclides (and other contaminants) can be transported throughout Puerto Rico’s various environments. Additional radionuclide inorganic chemistry, grain size, organic matter estimates and inventories will provide further insight to determine the local capacity and ability of sediment particles to retain contaminants (e.g. depleted uranium, associated trace isotopes and heavy metals). Analytical chemistry and clay minera logy analyses should be undertaken to investigate additional radionuclide properties to determine the distributions of naturally occurring (Pb-210, Ra-226, and uranium isotope s) and artificially produced Pu-238, Pu239, Pu-240, Am-241, Np-237 and Tc-99 radionuc lides. Sedimentation and mixing rates should be calculated after determining downcore excess 210Pb activities. Knowing these rates will provide insight regard ing the geological features of sediments in the island of Vieques and how long it takes to sediment to settle. Additional radionuclide analysis should be conducted to determine the presence, distribution and concen tration of depleted uranium in Vieques aquatic sediments. Dete rmining the concentrations and distributions of these radionuclides will further assist the VN WR in their efforts to create suitable and effective practices that will help the advancement of their mission to conserve our natural resources The data from the BONUS Area identifie d potential risks associated with the radionuclides left in the ento mbed monolith left at the f acility. The low retention indicated that there is no leak of the ra dionuclides left in the entombed; radionuclides released during the accidents at the BONU S have been leached out from the local environment. Knowing that there is low re tention is a positive f eedback to thousand of

PAGE 105

92 surfers and consumers of the area. The downcore profiles are providing information on the land use and how radionuclides serve as tracers to identify remobilization and disturbance of the soil in which the power pl ant was constructed. The other aspect that we need to consider is the f act that global fallout is not a major contributor to the total inventory of Cs-137 and exPb-210. Findings at the BONUS facility area also confirm that the absence of clays that serve as sinks for radionuclides might be anot her aspect contributing to the low retention of radionuclides in the area. Organic matter therefore plays an important role in the retention of radionuclides in so ils; it might be possible to fi nd higher activity levels on the crops and vegetation growing nearby the BONUS facility. Th e coastal water of Rincon is home of hundreds of marine and a quatic species which migrate from the cold winter to give birth and nurs e their calves in the tropical region. Endangered whales and turtles utilize Dome beach and adjacent beaches as nursery and hatching grounds. Therefore, plant-to-soil-transfer analysis should be conducted. Geochemical analysis of El Verde area and the Espiritu Santo provides information on environmental impacts that infl uences the retention of Cs-137. Cs-137 is present at all sampled sites. There is a difference in activity level at each site. This is likely the result of el evation, weathering, and environmenta l influences such as erosion, tidal processes, as well as organic matter content. Once depos ited at the estuarine system, radionuclides may have been further transported by local currents and tides. The beach area nearby the estuary is mostly composed of sand. Sand does not retain Cs-137. This study yiel ds interesting data that can be used for better planning if more experiments such as those conducted during the Rainforest Project are undertaken

PAGE 106

93 in the future. All radionuclides released into the environment via tree injection and inoculation experiments must be present at some level in the soil and sediments today. Some of the radionuclides used had a short ha lf-life. Others, like Cs-137 possess a long half-life and a fast mobility in the substrate. Vegetation at the rainforest was not sampled to study radionuclide transfer from soil-to-plants. The soil-toplant transfer will be an important study to conduct in the near future in order to see if the radionuclides that persist in soils are carried to different tr ophic levels that affect food web dynamics. Studies conducted in the late 1960’s show ed increased exposure of Cs-137 within the environment. These earlier studies n eed to be followed up with contemporary experiments conducted in the same manner. Geochemical data obtained through this research could be used to inform policy de cision in the eventuality that others would propose to conduct future experiments similar to the magnitude of Ra inforest Project. The absence of significant am ounts of Cs-137 might be due to the fact that the sample area did not have sufficient bind ing sites for the pollutant released. The activity levels found at the rainforest today might be the resu lt of atmospheric fall out resulting after the Chernobyl accident in 1986. As shown by Alonso-Hernandez (2002, 2006) the amount of Cs-137 is directly related to precipitation in Cuba and this might be the driving force in the rainforest. This statement is made afte r observing that the current Cs-137 activity in the rainforest does not represent the activity levels used by researchers during the RFP in the late 1960’s Poor retention can be cau sed by the absence of binding sites for the released pollutants. As described in the previous text, Cs137 retention on the island of Puerto Rico has been low and the main causes of this event in the ecosystem might be the variable

PAGE 107

94 sedimentation rate that has been cited by ot her researchers in the past as well as the absence of clay minerals. Clay minerals play a very important role in the retention and transportation of radionuclides, and on the island we only were able to document the presence of smectite and kaolinite in relatively low distribution in the soil and sediments.

PAGE 108


PAGE 109

96 INTRODUCTION Population increases and urban developmen t in coastal areas have impacted the aquatic and terrestrial environment in many ways. An example is the worldwide introduction of heavy metals from industria l and pharmaceutical activ ities (Critter and Airoldi, 2003, Bradl, 2004, Acevedo-Figueroa et al. 2006, Laing et al. 2009, Dong et al. 2009). Although many naturally oc curring metals are vital for biological processes, anthropogenic activities can increase metal concen trations to levels that are deleterious. In some instances impacts can even exte nd to alteration of ecosystem dynamics. Subsequent to the releas e of metals into the environment at elevated concentrations, their distributions, fluxes and potential for incorporation in the food chain can be influenced by many processes includ ing aqueous chemical speciation, sorption processes and bioturbation of contaminated soils and sediments (Asci et al. 2007). One major undesirable outcome of heavy meta l releases to the environment is bioaccumulation and biomagnification in the food chain. Consequently, the role of metals in food web dynamics and their mobility has become an increasingly important area of research in terms of environmental stew ardship, human health, and economics (e.g., DiGiacomo, et al. 2004, Valiela & Bowen, 2002, Shri ver, et al. 2002, Paul, et al. 2000). Heavy divalent metals such as Pb, Cu and Cd are common in nature at low concentrations. At sufficientl y high concentrations these meta ls can exert harmful effects on the ecosystem. Although Pb is found in the Earth’s crust at generally small

PAGE 110

97 concentrations (Manahan, 2005), after the industrial revoluti on its use increased sharply due to mining, burning of fossil fuel, and a va riety of manufacturi ng processes (Manahan, 2005). One major source of environmental pollution, leaded gasoline, was banned in many countries around 1995 (US EPA, 1996), and use of Pb in paint and other domestic products has also been reduced. Nevertheless, elevated lead levels in the environment continue to be a major concern (Manaha n, 2005). Unlike lead, Cu is a biologically essential element for many organisms (Van Genderen et al. 2005). At sufficiently elevated concentrations, however, dissolved c opper is toxic to unicel lular organisms (Van Genderen et al. 2005). Anthropogenic concentrat ions of Cu in the environment are the result of mining (O’Brien, W. 1997, Manaha n, 2005), use as a cladding on ships to reduce biofouling (Manahan, 2005) and the rel ease of industrial effluents (Manahan, 2005). Cadmium, another heavy metal that has become problematic due to anthropogenic activities, is found in the Ea rth’s crust at concentrations somewhat below the natural levels of Pb and Cu. It is concentrated and extracted during the production of other metals (O’Brien, W. 1997) and released into the ecosystem through municipal and industrial effluents (Manaham, S. 2005). Cd was once considered to have no known biological function in organisms (Van Genderen et al. 2005). However, at low concentrations, it provides some useful bi ological functions in a limited number of organisms (Van Genderen et al. 2005). Monovalent elements can also become pr oblematic in the environment due to human activities (Wyttenbach et al. 1995). Cs is one notable example in the form of released fission products, Cs-137 and Cs134, from the nuclear industry (Avery, 1996; Aakrog, 1994). Cs-137 was distri buted worldwide as a result of the Chernobyl nuclear

PAGE 111

98 accident in 1986. Subsequent to release into the environment, the propagation of elements is strongly influenced by interactions with particle surfaces (A lonso-Hernandez, 2006, Aarkrog, 2003, Moon, et. al, 2003, Livingst on & Povinec, 2000, Avery, 1996, Santchi and Honeyman, 1989). As such, an understanding of metal interactions with surfaces is essential to an understanding of me tal mobility in the environment. Sorption processes in sediments and pore waters can both concentrate metals and retard their movement through the ecosyste m (Fan et al. 2009, Morais-Barros et al. 2007). A variety of investigations has s hown that sorption processes are strongly influenced by pH and ionic strength (B ruemmer and Herms, 1986, Bradl, 2003, Critter and Airoldi, 2003, Morais-Barros et al. 2007, Fa n et al. 2009,). Metals binding to the surface of clay minerals can be especially im portant to the mobility of metals in the environment due to the high sorption capacity of clays and their ch aracteristically high surface area and high surface charge (Missana et al. 2008). Parameters of special importance to cation – clay inte ractions include temperature, pH, particle mass to surface area ratio, ionic strength, mineral type and porosity. Sorption is strongly promoted at high pH (Critter and Airoldi, 2003, Echevarria et al. 2005, Fan et al. 2009, Du Lai ng et al. 2009), and sorp tion decreases with increasing ionic strength (Echevar ria et al. 2005). Soil type or clay mineral type also affects metal sorption because mineral struct ures exert a strong influence on mineral reactivity. Clays that have a double layer stru cture are especially strongly sorptive. Porosity also plays an important role in so rption due to its influe nce on water percolation through the clay matrix, and t hus the extent to which wate r and clay surfaces can be brought into contact (Bruemmer et al. 1986).

PAGE 112

99 The aim of this project is to examine the influence of pH on metal sorption by both pure mineral substrates and natural soils. The pure substrates used in this work are illite, smectite (montmorillonite) and kaolinite (k aolin). The natural substrates used in this work are soil and sediment samples from th ree locations in the isla nd of Puerto Rico. This project constitutes the first examina tion of the influence of pH on trace metal sorption by soils from Puerto Rico. Although a few studies have described the presence of heavy metals in tropical ecosystems, incl uding estuarine and riverine environments in Puerto Rico, none have described the infl uence of pH on sorption by tropical soils (Acevedo-Figueroa et al. 2006, Hunter and Arbona, 1995, Infante and Acosta, 1991).

PAGE 113

100 METHODS Smectite and illite are double layer type cl ays and kaolinite is a single layer type mineral. Smectite and illite consist of two tetrahedral sheets. These sheets have a negative charge that is compensated by the ad sorption of cations. The interstitial space on clay minerals, specifically illite and smectite is distinctive (table 8). The interstitial space defines the size of the ions with whic h each mineral will preferentially bind. In some instances the larger interstitial spaci ng of smectite enables favorable binding with relatively large ions such as potassium. Kaolinite has a smaller capacity for binding available cations. The mineralogy of these cl ays is especially effective promoting the binding of metals such as K, Na and Cs. Table 8. Cation exchange capac ity (CEC) of clay minerals (Clay Minerals Society). Mineral Interstitial layer CEC Surface Area Kaolinite Only one surface plate 2.0 meq/100g 10.05 +/0.02 m2/g Illite 10 ngstrm 17 meq/100g 163 m2/g Smectite Between 10 and 18 ngstrms 76.4 meq/100g 31.82 +/0.22 m2/g Cation binding capacity or cat ion exchange capacity of minerals (CEC) is highly variable. The CEC for the common minerals, il lite, smectite and kaolinite, are presented in table 9. It is also the case that the intera ction of natural soils with different cations is highly variable. Soils are heterogeneous natura l systems composed of interacting organic and inorganic compounds. They are the final product of physical, chemical and biological weathering processes. The organic portion of soils comes from the decay of plants.

PAGE 114

101 Weathering processes ca n create complex soil horizons, and compositions that strongly vary with depth (Manahan, 2005). The natural soils and sediments used in th is work were collected in Puerto Rico (figure 47). Soil sediment cores were coll ected from the area adjacent to the BONUS Plant in western Puerto Rico, and from the El Verde Experimental Station in eastern Puerto Rico. Marine sediment samples we re collected from Ba hia Mosquito, Vieques and from the Espiritu Santo Estuary (figur e 47). Samples for these experiments were collected as described in Chapter 2 of this dissertation. Soils and sediments from bottom and top portions of the core were used to analyze any possible difference in the adsorption process as a function of core de pth. The depth of the cores in soils and sediments ranged from 20cm to 62cm. Figure 47. Sampling locations in Puerto Rico Chemical treatments were performed in a class 100 laminar flow clean air facility (manufactured by Laminaire). Trace metal clean water (Milli-Q water) was obtained using a Millipore purification system (Bed ford, MA). Teflon and polypropylene bottles,

PAGE 115

102 and polycarbonate filter me mbranes, were cleaned by soaking in HCl or HNO3 for at least 1 week and then rinsing several times with Milli-Q water prior to use. For all experiments, a pH 6.0 buffer so lution was prepared from 50 mM 2-(nmorpholino) ethanesu lfonic acid (MES, 99.5%, Sigma Aldrich) and 0.5 M NaOH. Trace Metal Grade nitric acid and certified 0.5M sodium hydroxide were purchased from Fisher Scientific (Pittsburg, PA). A metal stock solution in 2% HNO3, containing 16.67 ppm of each metal, was prepared from si ngle-element ICP standards (SPEX CertiPrep, Metuchen, NJ). A pH standard (pH 3.0, 0.001 M HCl) was prepared and used to calibrate the electrode. The pH of the experimental so lutions was expressed on the free hydrogen ion concentration scale. Measurements were obtained using a Ross-t ype combination pH electrode (No. 810200) connected to a Corning 130 pH meter in the absolute millivolt mode. Linearity and Nernstaian behavior of the electrode were veri fied by titrating a 0.3 M NaCl solution with concentrated HCl. Experimental Methods Three types of experiments were perform ed. In the first type of experiment, desorption of metals from minerals and soils was observed in the absence of any deliberate addition of metals to the expe rimental system. In the second type of experiment, rates of metal sorption by clay minerals were studied at approximately constant pH (pH~6) subsequent to additi on of monovalent (Cs and Rb) and divalent metals (Pb, Cu, Cd) to the experimental system .. In the third type of experiment, the pH dependence of distribution coefficients (so lid phase concentrati ons vs. solution phase

PAGE 116

103 concentrations) were determined using clay minerals (illite, kaolin, montmorillonite), terrestrials soils, and ocean sediments. Desorption Experiments Desorption experiments were performed over a range of pH. Solutions consisted of 5 mM MES at pH 6 added to the different types of substrates and 30l of internal standard. Incremental a dditions of either 2% HNO3 or 0.5 M NaOH were used to adjust the pH of the experimental solutions. An initi al sample (5ml) of the solution was used to determine the initial concentra tions of metals in each aque ous system. A 1:10 solid to solution ratio was achieved by addi ng 5 grams of substrate (clays or soils) to 50 ml of the aqueous solution housed in a centrifuge tube. Experiments were conducted at room temp erature t = 21 0.5 C. Samples were shaken for 24 hours on a (Eberbach) shaking table. Next, twenty-four experimental samples were centrifuged at 2000 RPM for 15 minutes in a HN-S Centrifuge (International Equipment Company). Followi ng centrifugation, the fi nal solution pH was measured. Two samples (5 ml each) were collected from the supernatant in the 50 ml vials for filtration. Five milliliters were used to rinse the polypropylene syringe and the Nuclepore filter (polycarbonate, 0.10 m pore size ). The syringe and the filter were then mounted on a polypropylene filter holder, a nd the second 5 ml from the centrifuged sample was filtered and collected in anothe r 50 ml centrifuge tube. Samples from the desorption experiment were acidified with 10l of concentrated HNO3. Thirty microliters of an internal standard solution containing e qual concentrations of In and Bi were then added to 5 milliliters of each acidified experimental solution.

PAGE 117

104 Sorption Kinetics Sorption kinetics experiments were perfor med by measuring the extent of metal sorption through time. Five hundred milliliter solutions consisting of 0.01 M ammonium nitrate, 5 mM MES (pH 6) and 100 ppb of each metal were housed. in 1L Teflon bottles An initial solution sample was collected prior to addition of 10g of solid (illite, kaolinite or montmorillonite). The contents of each bottle were stirred thr oughout each experiment using a (Thermolyne) magnetically-coupled stirring apparatus. Solution pH was measured and samples were taken at 10, 30, 60, 90, 120, 180, 360 minutes, 24 hrs and one week after initia tion of the experiment. Two 5m l samples were taken following a protocol essentially identical to that used in the desorption experiments. However, for measurements of sorption kinetics samples we re filtered but not centrifuged after each time increment. Samples were diluted 10-fold with 2% HNO3, and 50l of an internal standard solution containing equal concentrations of In and Bi was added to the solutions prior to analysis Sorption Experiments Sorption experiments were performed ove r a range of pH. Solutions contained 0.01 M using ammonium nitrate (99.999%, Sigm a-Aldrich), 5 mM MES, and 100 ppb of each metal. The overall experimental protoc ol, sample preparation and processing was similar to that used in desorption expe riments. The only dis tinction between the desorption and sorption experiments was the addition of metals in the latter. The concentrations of individual elements ranged between 0.48 and 1.17 M depending on each element’s atomic mass (see table 9). The sum concentration of all added elements in

PAGE 118

105 each experimental system was 4.87 M. Prior to analysis, samples were diluted 10-fold with 2% HNO3, and 50l of an internal standa rd solution that contained equal concentrations of In and Bi was added. Table 9. Initial concentra tion of metals added to adsorption experiment Metal Initial concentration Pb 483.6 nM Cu 1.57 M Cd 889.6 nM Cs 752.4 nM Rb 1.170 M Analysis Acidified samples were analyzed with an Agilent Technologies 7500cx inductively-coupled plasma mass spectromete r (ICP-MS). Solutions were introduced into the ICP-MS with a Micr o Mist concentric nebulizer and a double-pass (Scott-type) quartz spray chamber that was Peltier-cooled to t = 2C. During instrument tuning, the formation of oxide and double-charged ions was minimized with a 1 ppb Ce solution. MO+ and M2+ peaks were always less than 1.5% and 2% of the corresponding M+ peak, respectively, and correction fo r this effect proved unnecessary Metal concentrations were calculated from linear regressi ons of five standards (0.5, 1, 2, 5, and 10 ppb). A 2% HNO3 solution was run before and after the calib ration line, to serve as a blank and to rinse the instrument after the highest standard. In a ddition, after each autosampler position, Milli-Q water was aspira ted for 5 s followed by a 2% HNO3 wash solution for 60 s, in order to rinse the outside of the autosampler probe and the sample introduction system. All standards and solutio ns were injected in triplicat e. Ion counts were corrected for minor instrument drift by normalizing 65Cu, 85Rb, 111Cd, 114Cd, and 133Cs to 115In and

PAGE 119

106 normalizing 208Pb to 209Bi. Detection limits, calculated as three times the standard deviation of 37 blanks, were estimated to be 0.01 ppb for Cd, 0.03 ppb for Cs, 0.06 ppb for Rb, 0.08 ppb for Cu, and 0.09 ppb for Pb. Distribution coefficients were calculated by using the following equation: Ks = [MS]/[M]final Where [MS] = [M]T – [M]final = the concentration of sorbed metal and [M]T = [M]initial + [M]desorbed = the total concentration of metal in each experiment. [M]final = the final dissolved concentrati on of metal at the end of each experiment. [M]initial = the initial dissolved concentration of added metal in each experiment. [M]desorbed = the concentration of metal re leased from a substrate during desorption experiments.

PAGE 120

107 RESULTS Desorption experiments (clay minerals) Results for the desorption of referenc e materials are shown in Figure 48 and summarized in Table 10. Desorption behaviors we re distinctive for all 3 types of clays. Cu and Rb are extensively desorbed from all clays. Kaolin and montmorillonite produced relatively low concentrations of Cu and Rb ( 200 nM), while illite produced Rb concentrations on the order of 2, 500 nM. Rb concentrations produced by desorption from illite are more than twice as large as initial Rb concentrations produced by deliberate metal additions in sorption experime nts (table 10). Illite produced moderate concentrations of Cs (~ 225 nM), whereas for kaolin and montmorillonite, released Cs concentrations are relatively low (<11nM). Pb and Cd show the smallest extent of desorption from all clays (<2nM). Figure 48 also shows, for all metals, that the extent of desorption is not strongly dependent on pH. Metal desorption experiments using soils a nd sediments from Puerto Rico (figure 49 and table 12) show results that are simila r to those obtained using pure clay minerals (figure 48 and table 11). In general high desorbed concentrations of Cu ( 660 nM) were observed at all study sites. High desorbed concen trations of Rb were observed at all sites except for the El Verde area where desorption of Rb is 120 nM. Pb, Cd and Cs exhibit relatively minor extents of desorption ( 10 nM) from soils at all of the study sites. The

PAGE 121

108 soils from El Verde are the only samples that show a decrease in desorption with increasing pH (figure 49). Desorption at all ot her sites appears to be independent of pH. Desorbed concentrations of Cu and Rb vary from site to site but have concentrations generally similar to those produced by desorption fr om montmorillonite and kaolin. Desorbed Cs concentrations we re generally low for all minerals, soils and sediments with the exception of Illite, which produced Cs concentrations one to two orders of magnitude higher than was observed for other solids. Desorbed Cd concentrations were low for all solids, but we re generally higher for soils and sediments than was observed for clays minerals. Pb con centrations were low for clay minerals and generally only slightly highe r in soils and sediments.

PAGE 122

109 KaolinpH 456789 Log concentration (molar) -11 -10 -9 -8 -7 -6 -5 Cu Rb Cd Cs Pb Montmorillinite 456789 -11 -10 -9 -8 -7 -6 -5 Illite 456789 -11 -10 -9 -8 -7 -6 -5 Log concentration (molar) Log concentration (molar)pH pHRb Cu Cs Cd Pb Rb Cu Cs Cd Pb Rb Cu Cs Cd Pb Figure 48. Desoption experiment in clay minerals.

PAGE 123

110 Table 10. Summary of results of desorp tion experiments on clay minerals Element Desorption on Illite Average concentration (nM) Standard Deviation Highest concentration (nM) Lowest concentration (nM) Pb 0.40 0.23 0.55 0.11 Cu 62.60 42.82 140.26 28.29 Cd 0.40 0.20 0.77 0.23 Cs 225.20 21.94 249.88 198.17 Rb 2376.7 209.7 2630.2 2006.60 Element Desorption on Montmorillonite Average concentration (nM) Standard Deviation Highest concentration (nM) Lowest concentration (nM) Pb 0.23 0.24 0.62 0.04 Cu 22.97 5.60 32.87 17.40 Cd 0.94 0.72 1.98 0.31 Cs 6.70 2.44 10.62 3.436289 Rb 157.34 44.66 216.10 88.88 Element Desorption on Kaolin Average concentration (nM) Standard Deviation Highest concentration (nM) Lowest concentration (nM) Pb 0.24 0.18 0.60 0.04 Cu 25.54 17.95 59.70 11.40 Cd 1.15 0.66 2.05 0.16 Cs 2.84 1.10 5.11 1.57 Rb 19.31 7.85 35.84 10.27

PAGE 124

111 CopperpH Log concentration (m olar) -8.0 -7.5 -7.0 -6.5 -6.0 Espiritu Santo El Verde Rincn Vieques CadmiumpH Log concentration (m olar) -10.5 -10.0 -9.5 -9.0 -8.5 -8.0 LeadpH Log concentration (m olar) -10.5 -10.0 -9.5 -9.0 -8.5 -8.0 RubidiumpH Log concentration (m olar) -8.0 -7.5 -7.0 -6.5 -6.0 CesiumpH Log concentration (m olar) -10.5 -10.0 -9.5 -9.0 -8.5 -8.0 EV ES V R EV ES V R EV R V ES EV ES R V EV ES R V Figure 49. Desorption experiments on soils and sediments from Puerto Rico

PAGE 125

112 Table 11. Summary of results of desorption experiments for soils and sediments in Puerto Rico. Element Desorption at Espiritu Santo Average concentration (nM) Standard Deviation Higher concentration (nM) Lower concentration (nM) Pb 0.56 0.33 1.34 0.29 Cu 253.67 171.55 659.36 101.14 Cd 3.95 2.79 8.87 1.77 Cs 0.41 0.12 0.65 0.28 Rb 194.94 81.88 342.11 90.83 Element Desorption at El Verde Exp. Station Average concentration (nM) Standard Deviation Higher concentration (nM) Lower concentration (nM) Pb 0.75 0.50 7.91 0.09 Cu 269.16 58.73 370.13 180.34 Cd 2.24 1.78 5.41 0.14 Cs 2.47 1.08 4.58 1.41 Rb 55.73 26.49 121.33 24.89 Element Desorption at BONUS Area Rincon Average concentration (nM) Standard Deviation Higher concentration (nM) Lower concentration (nM) Pb 1.97 3.20 7.90 0.03 Cu 84.19 13.74 105.12 57.89 Cd 1.24 0.73 2.29 0.35 Cs 1.05 0.36 1.70 0.53 Rb 272.29 59.79 341.53 126.93 Element Desorption at Vieques Average concentration (nM) Standard Deviation Higher concentration (nM) Lower concentration (nM) Pb 0.39 0.27 0.86 0.15 Cu 136.33 93.61 308.91 40.00 Cd 3.094 2.00 6.92 0.83 Cs 1.19 0.35 1.63 0.85 Rb 243.59 110.48 500.66 130.70

PAGE 126

113 Equilibration kinetics (clay minerals) Several sorption experiment s were performed over time (Figure 50) at constant pH (~6) to determine an optimum sampling period for sorption experiments that were performed over a range of pH. These expe riments, conducted using the clay reference materials, showed that equilibrium between th e solutions and all clay minerals (illite, kaolin and montmorillonite) was attained on short time scales. Only small changes in concentration were observed after 120 minutes (figure 50) The equilibria between sorbed and dissolved metals can be summarized as follows: Pb and Cu were sorbed to the greatest extent, Cd was sorbed to a lesser exte nt, and Cs and Rb were adsorbed to much lower extents than the divalent elements. The solution concentrations of all metals except Rb decreased sharply within the first 10 minutes of each experiment. For Rb this decrease was observed only in the case of Montmorillonite. Due to strong desorption, Rb concentrations actually increased through time

PAGE 127

114 KaolinTime 0100200300400 Log concentration (molar) -9.0 -8.5 -8.0 -7.5 -7.0 -6.5 -6.0 -5.5 Cu Rb Cd Cs Pb IlliteTime 0100200300400 Log concentration (molar) -9.0 -8.5 -8.0 -7.5 -7.0 -6.5 -6.0 -5.5 MontmorilloniteTime 0100200300400 Log concentration (molar) -9.0 -8.5 -8.0 -7.5 -7.0 -6.5 -6.0 -5.5 Rb Cs Cd Cu Pb Rb Rb Cs Cs Cd Cd Cu Cu Pb Pb Figure 50. Element concentration over time on clay minerals

PAGE 128

115 Sorption on reference material Figure 51 shows results clay minerals adsorption experiments performed using illite, kaolin and montmorilloni te, and Table 12 provides log Ks averages and standard deviations Figure 51 demons trates that the extent of adsorption is not strongly dependent on pH for the metals and clays us ed in this study. Di stribution coefficient values ranged over two to three orders of magnitude. The highest average log Ks values were observed for Pb (log Ks (Pb) ~ 2.7) and the lowest values were observed for Rb on Illite (log Ks (Rb) = -0.55). Observed average log Ks values had the following order Pb > Cu > Cd > Cs > Rb. The average distribution coefficients obser ved (Table 13) for two of the divalent cations, Pb and Cu, were quite consistent for all clay mineral substrates ((log Ks (Pb) ~ 2.7 and log Ks (Cu) ~ 1.6). For the remaining elemen ts, distribution coefficients were increasingly variable between the substrates, an d the extent of variability increased in the order Cd < Cs < Rb. Average distribution co efficients for Cd were as low as log Ks (Cd) ~ 0.7 (kaolin) and as high as log Ks (Cs) ~ 1.3 (montmorillonite). Cs and Rb log Ks averages were even more variable between th e three types of clay minerals: Average log Ks values for Cs were as low as 0.27 (kaolin) and as high as 1.5 (montmorillonite) Average log Ks values for Rb ranged between -0.55 (illite) and 0.73 (montmorillonite). For Cd, Cs and Rb, the highest average distribution coefficients were always observed for montmorillonite. The lowest averag e distribution coefficients for Cd and Cs were observed for kaolin and the lowest log Ks value for Rb was obtained for illite, the clay mineral that produced net desorption in the Figure 50 evaluation of sorption kinetics.

PAGE 129

116 Illite pH 5678 Log Kd -1 0 1 2 3 4 Cu Rb Cd Cs Pb Kaolin pH 5678 Log Kd -1 0 1 2 3 4 MontmorillitepH 5678 Log Kd -1 0 1 2 3 4 Pb Cu Cd Cs Rb Pb Rb Cs Cd Cu Pb Cu Cd Cs Rb Figure 51. Metal adsorption on clay mineral

PAGE 130

117 Table 12. Summary of results of adsorp tion experiments on clay minerals Element Adsorption Montmorillonite Average log Ks Standard Deviation Highest value Lowest value Pb 2.70 0.42 3.27 2.00 Cu 1.67 0.26 2.00 1.39 Cd 1.33 0.24 1.70 1.00 Cs 1.50 0.13 1.63 1.30 Rb 0.73 0.16 0.90 0.53 Element Adsorption Kaolin Average log Ks Standard Deviation Highest value Lowest value Pb 2.70 0.56 3.30 1.81 Cu 1.70 0.19 2.04 1.40 Cd 0.68 0.46 1.30 0.20 Cs 0.27 0.10 0.44 0.12 Rb -0.20 0.15 0.04 -0.43 Element Adsorption Illite Average log KsStandard Deviation Highest value Lowest value Pb 2.70 0.39 3.06 2.19 Cu 1.53 0.098 1.67 1.37 Cd 1.03 0.24 1.42 0.71 Cs 0.45 0.07 0.55 0.33 Rb -0.55 0.41 0.02 -1.16

PAGE 131

118 Sorption on soils and se diments in Puerto Rico Metal sorption experiments on soils and sedi ments from the island of Puerto Rico are shown in figure 52. Table 15 shows sorp tion comparisons for all four soils and sediments at the study sites. As was observe d using the pure clay minerals, the most strongly sorbed metal was Pb and the least stro ngly sorbed metal was Rb. In contrast to the order of log Ks.values observed for the pure clay minerals, Cd was sorbed more strongly than any metal other than Pb, and Cs was sorbed more strongly than Cu at all sites except the BONUS site at Rincon. As in the case of the experiments perfo rmed using pure clay experiments, pH adjustments were performed in the experiment s with soils and sediments. The pH ranges in these experiments are shown in Table 14. Soils from El Verde Experimental Station (EV) had the lowest pH of all samples, st arting with a pH of 4.4. Even though the soil from El Verde ES was treated with 0.5 M NaOH to gradually increase the pH, the treatment only increased pH by 1.4 units to a maximum value of 5.8. All soils had a substantial buffer capacity, similar to that seen in experiments with illite. Samples from Rincon exhibited a change of only 0.6 pH units after addition of 2% HNO3. The pH of sediments from the Espiritu Santo River and Mosquito Bay (Vieques) changed by only 0.7 units and 0.6 units after similar treatments. For all soil types except El Verde, there was no easily discernable trend of log Ks values with pH. For the acidic El Verde soils, log Ks generally increased with increasing pH. The strongest increase was observed for Cd sorption, and only in the case of Cu wa s there no observed trend at this site. Results from the clay mineral experime nts can be compared and contrasted with the results obtained using soils and sediments. Distribution coefficien ts observed for Pb

PAGE 132

119 in both pure clay minerals and the natural so ils were closely comparable. Average log Ks values obtained with soils and sediments ranged between 2.45 a nd 2.84 compared to an average of 2.70 for the pure clay minera ls. In the case of Cu, Average log Ks values obtained using soils and sediments range d between 0.56 and 1.12. These values are substantially smaller than th e range of values observed us ing pure clay minerals (1.53 log Ks 1.70. In contrast to the lower log Ks values for Cu observed using natural samples, average log Ks (Cd) values obtained for natural soils and sediments (1.11 log Ks(Cd) 2.02) are substantially higher that t hose observed using pure clay sediments (0.68 log Ks(Cd) 1.33). Distribution coefficients for Cs and Rb obtained using natural samples were broadly comparable to those observed using pure minerals: 1.05 log Ks(Cs) 1.58) vs. 0.27 log Ks(Cs) 1.50) and -0.50 log Ks(Rb) 0.42) vs. -0.55 log Ks(Rb) 0.73) Table 13. Experimental pH for each soil type Sample Lowest pH Highest pH Sample type Mosquito Bay, Vieques 6.85 7.60 Ocean Sediment Espiritu Santo, Rio Grande 6.3 7.12 Ocean Sediment El Verde Experimental Station, Rio Grande 4.36 5.81 Soil Area adjacent to BONUS Power Plant, Rincon 6.66 7.22 Soil

PAGE 133

120 CopperpH Log K d -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 Espiritu Santo El Verde Rincn Vieques RubidiumpH Log k d -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 CadmiumpH Log K d 0.0 0.5 1.0 1.5 2.0 2.5 3.0 CesiumpH Log K d -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 LeadpH Log K d 1.5 2.0 2.5 3.0 3.5 4.0 4.5 EV ES R V EV ES V R EV ES V R EV ES V R EV ES V R Figure 52. Metal behavior in soils and sediments from different study sites.

PAGE 134

121 Table 14. Summary of result of adsorption expe riments for soils and sediments in Puerto Rico. Element Adsorption at El Verde Exp. Station Average log Ks Standard Deviation Highest value Lowest value Pb 2.45 0.45 3.00 1.82 Cu 0.56 0.14 0.71 0.38 Cd 1.11 0.58 1.88 0.33 Cs 1.07 0.17 1.34 0.79 Rb 0.42 0.19 0.76 0.17 Element Adsorption at BONUS Area Rincon Average log Ks Standard Deviation Highest value Lowest value Pb 2.84 0.64 3.77 2.00 Cu 1.12 0.17 1.27 0.70 Cd 1.89 0.17 2.19 1.66 Cs 1.05 0.37 1.42 0.45 Rb -0.14 0.10 0.06 -0.29 Element Adsorption at Espiritu Santo Average log Ks Standard Deviation Highest value Lowest value Pb 2.68 0.41 3.38 2.10 Cu 0.62 0.16 0.80 0.40 Cd 1.85 0.22 2.30 1.59 Cs 1.58 0.04 1.68 1.54 Rb 0.07 0.19 0.53 -0.001 Element Adsorption at Vieques Average log Ks Standard Deviation Highest value Lowest value Pb 2.84 0.69 4.09 2.16 Cu 0.92 0.23 1.22 0.62 Cd 2.03 0.35 2.57 1.64 Cs 1.11 0.19 1.37 0.85 Rb -0.50 0.10 -0.68 -0.33

PAGE 135

122 DISCUSSION The soils and sediments from Puerto Rico do not contain illite at any study sites. However, kaolinite and smectite are present at all stations. Samples from the BONUS site are approximately 20% clays as a mixture of kaolinite and smectite. Soil at El Verde Experimental Station consists of approximate ly 15% clay, principally in the form of kaolinite. Sediment at the Espiritu Santo Es tuary is approximately 10% clay in the form of smectite and kaolinite. An understanding of soil composition is important to interpretation of sorption properties with respec t to the contributions of the clay minerals studied in this investigation. Samples from the BONUS site have bot h comparatively high Ks values and comparatively high clay content. Samples fr om other sites in the island have similar fractions of smectite and kaolinite. Clay minera logy in Puerto Rico is a likely controlling factor for metal mobility. Based on observed Ks values, Pb is likely to have a comparatively slow mobility and high retenti on in the soils and sediments of Puerto Rico. Accordingly, the mobility of Cu, Cd, Cs, Cd a nd Rb will be substantially faster than Pb because the lower log Ks of these metals.. Soil pH does not strongly affect th e sorption of metals for most of the investigated clays, soils a nd sediments. This trend is generally observed except for Kaolin and the acidic soils from El Verd e Experimental station. The influence of increasing pH, on the log Ks of Cd is es pecially notable. The observations of

PAGE 136

123 comparative Pb and Cd sorption obtained in th is work are consistent with those of Appel and Ma (2002) who reported that a variety of tropical soils types, Oxisol, Ultisol and Mollisol, have a preference for Pb sorption rela tive to Cd. The Ks values for Cu sorption by soils are smaller than those obtained in th e clay mineral experiments. This behavior might be due to the nature of Cu complexation and speciation in sediments. Delgadillo et al. 2008, states that up to 80% of the tota l dissolved Cu in surface waters can be organically complexed. As such, its sorpti on availability woul d be reduced. This explanation for the behavior of Cu in soil sorption studies would re quire that organics present in the soils were released into solu tion during the sorption e xperiments. This is consistent with the observation that the sa mples demonstrating unusual adsorption of Cu are the samples from the comparatively organicrich rainforest ecosystem and its estuary. The results obtained in sorption experime nts help in interpreting comparative retention of metals and radionuclides in so ils and sediments from Puerto Rico. Pb-210 concentrations in most cases are higher th an Cs-137 concentrations. Since distribution coefficients are highest for Pb for all soils and sediments, it expected that Pb-210 will not travel far in the sediment column before be ing sorbed by the substr ate. Cs distribution coefficients are much smaller. Therefore Cs is likely to be far more mobile in the soils and sediments in the island. After government al activities in which the amount of Cs-137 introduction was extremely large, the obser ved amounts of Cs-137 in the Puerto Rican environment today are comparable to the fallo ut Cs-137 contribution in this region of the hemisphere. Nevertheless, the soil com position of Puerto Rican soils can have a significant impact on Cs-137 environmental activ ity. At the rainforest study site where

PAGE 137

124 most samples are especially rich in smectite Cs-137 inventories are higher than at any other location within the area. This study furthers an understanding of geochemical processes in Puerto Rico through investigations at four different ecosystems on the island. Ecosystems such as the tropical rainforest receive more than 500 cm of rain a year, commonly causing floods and landslides. Puerto Rico is also one of th e primary areas where hurricanes develop and, on decadal scales, cause major ecosystem destru ction. The mobility and soil-retention of metals such as Pb, Cd, Cu, Cs and Rb examin ed in this study will help to provide an understanding of ecosystem interaction in these extreme, but natural events.

PAGE 138


PAGE 139

126 GEOCHEMISTRY OF THE STUDY SITES AND HOW THEY IMPACT CURRENT RESULTS Resource managers and policymakers ofte n rely on environmental studies and monitoring efforts conducted by researchers in order to implement effective manage practices designed to preserve natural re sources. Although hypotheses frequently guide scientific investigations, outcomes of some investigations can be surprising and do not always support the original hypothesis. For this environmental assessment project, the working hypothesis was that Cs-137 activity of soil and sediments would be elevated within three study sites in Puerto Rico as a result of local military and federal government activities. However, samples analyzed duri ng this study indicate that Cs-137 (and Pb210) activity does not appear to be particularly elevated at any of the three study sites. The sorption and retention of Cs-137 de pends on various geochemical properties and environmental conditions. Clay minera logy influences the sorption and retention ability of sediments and soils. Illite and smec tite, in particular, provide effective Cs-137 binding sites due to the intersti tial spaces of these two clay minerals. Particle size, water and in-situ chemistry also influence the ability of Cs-137 to be abso rbed and retained in sediments and soils (Johnson-Pyrtle and Sco tt, 2001). Results of this investigation indicate that, in general, ther e appear to be particles of suitable size for Cs-137 sorption and retention size throughout all three study sites (see figures 22-31). During this investigation it was dete rmined that clay mineralogy of the study sites, particularly the

PAGE 140

127 limited presence and/or absence of illite a nd smectite, contributed to the soil and sediment’s inability to effectivel y retain locally introduced Cs-137. Salinity also affects the retention of Cs-137. At higher salinities Cs-137 will be desorbed from fluvial sediments due to cat ion competition reactions (Avery, 1996). Of the various ecosystems sampled during this investigation only thr ee of the study sites, Mosquito Bay and Green Bay (Kiani Lagoon) at Vieques, and/or Espiritu Santo River area’s Coco Beach might have been aff ected by changes in porewater and surface salinity. All other samples examined during this investigation are either soils or were collected from freshwater systems. Conseque ntly, changes in salinity are not factors for absorption and retention of Cs-137 at the major ity of the sites that were examined during this investigation. Physical processes and events also influe nce the retention of Cs-137 in sediments and soils. In addition to experiencing majo r storms on an annual basis, and frequent hurricane events, various regions of Puerto Rico are also subj ect to flash floods, landslides and particle erosion. All of these natural physical processes, in addition to anthropogenic activity, have the ability to in fluence the retention of Cs-137 by removing and transporting Cs-137 laden particles away from their original introduction site, beyond the local study sites and potentially offshore. Cs-137 activity was expected to be highe r in the study sites based on the initial activity that was released at each site. As mentioned previously, the amount of Cs-137 that was irradiated at the El Verde Experimental Station was 1014 Bq and the estimated amount of Cs-137 that remains in the entomb ed monolith at the B ONUS facility is 2.088 x 107 Bq (see chapter 2 for details). Cs-137 has not been locally introduced at Vieques.

PAGE 141

128 As such it is anticipated that the sole source of Cs-137 pres ent in Vieques is from is global fallout. The Cs-137 activity of sample s collected from Vieques is comparable to activities determined for all of the other study sites, indicating that the majority of Cs-137 present on the island of Puerto Rico is the result of global atmospheric fallout. Furthermore, the average Cs-137 activity dete rmined for each of th e three study sites is comparable to Cs-137 activities in other regi ons of the Caribbean. The closest study (in terms of proximity to Puerto Rico) was pe rformed in Cienfuegos, Cuba. There AlonsoHernandez et al. (2006) indicated that the average atmospheric deposition of Cs-137 in the region was 14 dpm m-2. A radionuclide assessment study was also performed for the Costa Rican tropical ecosystem. This Co sta Rican investigation, which was conducted after the Chernobyl accident revealed that mean Cs-137 activity of 35,000 dpm/m-2 (Bossew and Strebl, 2001). LeBrecque and Cordoves (2001) found th at Cs-137 activities in tropical soils of two Venezuelan rainforests range from 120 to 600 dpm/m-2. Cs-137 activities were also determined for the Miami, Florida area. As previously mentioned in Chapter 2 (see figure 47), the Miami, Florida area received Cs-137 as a result of global atmospheric fallout. The Cs-137 activity peaks determined in the Miami, Florida area coincide with periods of nuclear weapons testing and the Chernobyl accident. Cs-137 deposition in the Miami, Florida area has decreased continuously following the Chernobyl accident. Cs-137 activity present in th e Caribbean appears to primarily be the result of global fallout. Although the amount of activit y in the region varies, it should be noted that the values determined during this invest igation are comparable to those that were determined elsewhere in the Caribbean, specifi cally in Cuba (Alonso-Hernandez, 2006).

PAGE 142

129 The variations identified throughout the Caribb ean are likely the result of environmental influences and geochemical properties. The average Cs-137 activity for the BONUS, Vieques and El Verde Experimental Sta tion sites were 0.94 dp m/g, 0.15 dpm/g and 2.93 dpm/g, respectively. This investigation represents the firs t time a radionuclide assessment project has been undertaken at multiple locations throughout the island of Puerto Rico. The fact that this investigation revealed the low retention of Cs-137 in local soils and sediments is an important research finding. Results presente d in Chapters 2 and 3 of this dissertation indicate that to some extent the local ecosystems have the ab ility to retain Cs-137 and Pb210. Findings from this investigation (see chap ter 3 for details) reveal that all soils and sediments collected during this study were able to retain Cs and Pb. (Actually Pb was the most common metal sorbed during the adsorption experiment.)

PAGE 143

130 MAJOR CONCLUSIONS The results of this investig ation indicate that the Cs137 present in Puerto Rican soils and sediments is the result of fallout occurring in the Caribb ean region. Highest Cs137 activity was found in surface sa mples collected from the rainforest at the El Verde Experimental Station. The higher elevation and precipitation at this st udy site as well as soil composition might be the contributing factor s to the activity in the site. This Cs-137 activity is similar to what was reported by Alonso-Hernandez et al. (2002, 2006). They conducted a series of experiments in Cienfue gos, Cuba in which they reported a direct relationship between precipitation and Cs-137 activity. In addition, it should also be noted that LaBrecque et al. (2001) found that Cs-137 activity was higher in the Venezuelan cloud forest compared to nearby st udy sites at lower el evations. The cloud forest is the highest ecosystem in the tropical rainforest and the el evation of El Verde experimental station is lo cated at a proximity close to the cloud forest. As shown in Chapter 3 of this dissertati on the distribution coe fficients for Cs and Pb are different and the soils of the ecosys tem favors the adsorption of Pb to a greater extent when compare with Cs. For most of the cores on the is land, Pb-210 has a higher activity than Cs-137. It might be proposed that Pb-210 input in the is land is higher than the input of Cs-137. Since Cs-137 is an an thropogenic radionuclide, and Pb-210 is a natural occurring radionuclide we will need to analyze for the balance in the uranium decay series, which produce Pb-210.

PAGE 144

131 It is well known that after the Cher nobyl accident, Cs137 regional fallout depended on wind force. Therefore anot her possible explanation to the minimum activities of Cs-137 in the island is that th e Caribbean region did not receive the same activity that was received in the mainland of the US and the European continent. We have evidence in this disserta tion that Cs-137 is present at low levels at all sampling locations, therefore the governmental activ ities and accidental release by the BONUS Power Plant had no effect on the curren t activity level in the island. In terms of mineralogy, the substrate provided suitable bind ing sites although illite is not present at any of the study sites. Smectite a nd kaolinite content at all study sites was relatively low when compared to th e amount of sand presen t at each site. High fine particle size content is related to hi gher Cs-137 retention potential. Sand does not retain radionuclides as efficiently. It has been demonstrated by many researchers that the presence of fine grain clay minerals is vi tal for Cs-137 binding in substrates. Illite formations depend on the weathering of rocks and the resulting residual soils. There are 215 soil types in Puerto Rico and all study sites are located in different geological formations. Therefore the absence of illite can be attribute to ecosystem characteristics. For instance, the BONUS area is located with in a limestone formation. pH range among the soils and sediments went from 4 to 8. Binding with other clay particles would be possible since there are other clay minerals su ch as vermiculite, saprolite, chlorite and muscovite but the binding would not be as st rong and Cs-137 would be released after its binding. Organic matter can also facilitate Cs-137 retention. The organic matter content at the study sites went from low to intermediate The higher organic matter content in the

PAGE 145

132 rainforest might be due to the degradation of leaves and debris. For instance, Arctic sediments, where intentional release of ra dionuclides also occurred, showed activity levels 10 times higher than what we found at the rainforest. Another factor we should take into consideration is the fact that acco rding to the ICP-OES analysis, the substrate was oxic, which implies that Cs-137 mobility is low in the sediments. Cation competition is another factor that adversel y influences the competition for binding sites among pollutants. At study sites, binding s eems to be low and the presence of potassium as well as other elements might outcompete Cs -137 for binding the analyzed substrate. Results indicate that although intentional release of radionucli des occurred in two of the study sites of Puerto Rico, Cs-137 activity is not residual activity from the governmental activities and is more likel y the result of globa l fallout. Due to environmental conditions, once introduced, a portion of the Cs-137 may have been transported away via erosion, as well as runoff, currents, and tide induced water movement. Fluctuations in Cs-137 downcore pr ofiles were the result of fluctuations in global fallout and local substrate disturban ce, such as remobilization, hurricanes and other atmospheric events as well as bioturbation. At th e rainforest, high precipitation may have served to move the soils form one site to the other via a variety of particle movement mechanisms, including landslides that commonly occur in the area. The Rainforest Project (O dum and Drewry, 1970), constr uction of the prototype BONUS power plant and the Ammu nition facility in Vieques are notable events in the island’s history. However, Puerto Rico has been fortunate to have not experienced major disasters associated with these activities. After completing the BONUS reactor and El Verde experiments, no further studies were pref ormed to monitor the long term effects of

PAGE 146

133 the associated radio-pollution. Investigating the long-term effects of these experiments is vital to understand the effect of the introduction of radionuclid es into the ecosystem. Dr. Ariel Lugo, the director for the International Inst itute for Tropical Forestry, stated that the “irradiated forest was studied for twenty thr ee years, until 1988; at that time it had not fully recovered” (Intervi ew by Suzanna Engman, 2007). The current investigation provides radiogeoc hemical data that will be useful to policy makers in determining whether or not to authorize similar types of experiments in the future. Gamma spectroscopy has yielded information to analyze doses at which the human population could be exposed. Dose rates at the rainforest went from 5 r/hr (before the late 1960’s experiments) to 200 r/hr which represent an enormous increase that can adversely affect human health. In Vieques cleanup work is still in progress and finding the missing ammunitions is important. Pote ntial pollutants can be stored in these ammunitions and after their explosion sedime nts can serve as the sink for anthropogenic elements used during the military practices, further affecting human and environmental health. A program on open explosion has been recommended by the US Department of Defense and it is estimated that all related environmental clean-up work will last another 14 years. Information provided by this i nvestigation should prove beneficial when designing additional studies at the VNWR. One such study should be conducted at the target facility, which could not be accessed du ring this investigation due to its closure by the Nuclear Regulat ory Commission. An in-depth analysis of the erosion and sedimentation at the three study sites is recommended in order to further understand where the sediments are being transported. At the rainforest, a long-term study should be undertaken to analyze the soil to plant

PAGE 147

134 transfer factor in the area, and see if that is the cause of the slow growth of the vegetation located at the irradiated a nd injection areas (Odum and Dr ewry, 1970). An investigation of soil to plant transfer of radionuclides shoul d also be conducted at the BONUS facility area in order to determine the potential im pact of radio-polluti on in the event that radionuclides present at this f acility are accidentally released into the environment.

PAGE 148

135 REFERENCES Aarkrog, A. (2003) Input of anthropogenic ra dionuclides into the World Ocean. DeepSea Research II. 50, 2597–2606 Aarkrog, A. (1994) Radioactivity in Pola r Regions main sources. J. Environ. Radioactivity. 25, 21-35 Acevedo-Figueroa, D., B.D. Jimenez, and C.J. Rodriguez-Sierra. (2006) Trace metals in sediments of two estuarine lagoons from Puerto Rico. Environmental Pollution. 141, 336-342 Alonso-Hernandez, C.M., M. Diaz-Asencio, A. Munoz-Caravaca, R. Delfanti, C. Papucci, O. Ferretti, et al. (2006) Recen t changes in sedimentation regime in Cienfuegos Bay, Cuba, as inferred from Pb-210 and Cs-137 vertical profiles. Continental Shelf Research. 26:2, 153-167. Alonso-Hernndez, C.M., H. Cartas-guila, M. Daz-Asencio, A. Muoz-Caravaca, J. Martn-Prez, & R. Sibello-Hernndez. (2006) Atmospheric deposition of 137Cs between 1994 and 2002 at Cienfuegos, Cuba. Journal of Environmental Radioactivity. 88:2, 199-204. Alonso-Hernandez, C., M. Diaz-Asencio, A. Munos-Caravaca, E. Suarez-Morell, R. Avila-Moreno. (2002) 137Cs and 210Po dos e assessment from marine food in Cienfuegos Bay (Cuba). Journal of E nvironmental Radioactivity. 61:2, 203-211. Anoruo, A. O., C. H. Jagoe, J. I. Blake, F. C. Anoruo and J. D. Salley. (2002). Long-term effects of radiocesium (Cs-137) on forest growth and development. Journal of Environmental Monitoring a nd Restoration. 1(1):1-13

PAGE 149

136 Appel, C. and L. Ma (2002). Concentrati on, pH, and Surface Charge Effects on Cadmium and Lead Sorptionin Three Tropical Soils. Journal of Environmental Quality. 31, 581–589. Arapis, G.D. & M.G. Karandinos. (2004) Migra tion of Cs-137 in the soil of sloping seminatural ecosystems in Northern Greece. Journal of Environm ental Radioactivity. 77 (2), 133-142. A i, Y., M.Nurbas and Sa Aikel. (2007) Sorption of Cd(II) onto kaolin as a soil component and desorption of Cd(II) from ka olin using rhamnolipid biosurfactant, Journal of Hazardous Materials. B139 50–56. Augustin, J. and R. Zedja. (1991) Cancer incidence and geochemi cal factors in the environment. The Science of The Total Environment. 106:1-2, 155-163. Avery, S. (1996) Fate of cesium in the envir onment, distribution between the abiotic and biotic component of aquatic and terrestr ial ecosystem. Journa l of Environmental Radioactivity. 30 (2), 139-171 Ayala, C and V. Carro. 2005."Expropiations a nd Displacement of Civilians in Vieques, 1940-1950." In Puerto Rico under Colonial Rule: Political Persecution and the Quest for Human Rights, edited by R. Bos que Prez and J. Coln Morera Baqar, R., Zaidi, L. M. Hinkey, N. R. Rodriguez, Nadathur S., G. Syed & H. Imam. (2003) Biodegradation of toxic chemicals in Guayanilla Bay, Puerto Rico Marine Pollution Bulletin. 46, 418–423 Bradl, H. B. (2004). Adsorption of heavy metal ions on soils and soils constituents. Journal of Colloid and Interface Science. 277, 1–18. Bruemmer, G.W. J. Gerth and U. Herms. (1986) Heavy metal species, mobility and availability in soils. Journal of Plant and Soil Science. 149:4, 382-398 Bossew, P. and F. Strebl. (2001) Radioactive c ontamination of tropical rainforest soils in Southern Costa Rica. Journal of E nvironmental Radioactivity. 53:2, 199-213. Charmasson, S. (2003) Cs-137 inventory in sedi ment near the Rhone mouth: role played by different sources. Ocea nologica Acta. 26(4), 435-441.

PAGE 150

137 Ciffroy, P., G. Durrieu, and J.-M. Garnier. ( 2009) Probabilistic dist ribution coefficients (Kds) in freshwater for radioisotopes of Ag, Am, Ba, Be, Ce, Co, Cs, I, Mn, Pu, Ra, Ru, Sb, Sr and Th – implications for uncertainty analysis of models simulating the transport of radionuclides in rivers. Journal of Environmental Radioactivity. 100, 785–794. Claval, D., L.Garcia-Sanchez, J. Real, R. Rouxel, S. Mauger and L. Sellier. (2004) Erosion of atmospherically deposite d radionuclides as affected by soil disaggregation mechanisms. Journal of Environmental Radioactivity. 77, 47-61. Critter, S.A.M., and Claudio Airoldi. (2003) Adsorption–desorption processes of calcium on Brazilian soils. Geoderma. 111, 57–74. Dean, W.E., Jr. (1974) Determination of Ca rbonate and organic matter in calcareous sediments and sedimentary rocks by loss on ignition: Comparison with other methods. Journal of Sedi mentary Petrology. 44, 242-248. Desmarais, A. P. and B. He lmuth. (1970) A tropical rainfore st; A study of irradiation and Ecology at El Verde, Puerto Rico. Oak Ridge, TN: Atomic Energy Commission Daz, E. and A. Massol-Dey. 2003. Trace element composition in forage samples from military target range, three agricultural areas, and one natural area in Puerto Rico. Caribbean Journal of Sc ience. 39 (2), 58-63. Dong, D., X., Zhao, X. Hua, J. Liu, M. Ga o. (2009) Investigation of the potential mobility of Pb, Cd and Cr(VI) from mode rately contaminated farmland soil to groundwater in Northeast, China. Jour nal of Hazardous Materials. 162, 1261– 1268. Du Laing, G., J. Rinklebe, B. Vandecasteele, E. Meers, and F.M.G. Tack. (2009) Trace metal behavior in estuarine and riverine floodplain soils and sediments: A review. Science of the Total Environment. 40:7 3972-3985. Echeverria, J.C., I. Zarranz, J. Estella, and J. J. Garrido. ( 2005) Simultaneous effect of pH, temperature, ionic strength, and initial concentration on the retention of lead on illite. Applied Clay Science. 30, 103–115.

PAGE 151

138 Eisenbud, M and T. Gessel. (1997) Environm ental Radioactivity. San Diego CA: Academic Press. Fan, Q., Z. Li, H. Zhao, Z. Jia, J. Xu and W. Wu (2009) Adsorption of Pb(II) on palygorskite from aqueous solution: Effects of pH, ionic strength and temperature. Applied Clay Science. 45, 111–116. Fanning, K., R. Byrne, J. Breland II, P. Bet zer, W. S. Moore, R. Elsinger et al. (1981) Geothermal springs of the West Flor ida continental shelf: Evidence for dolomatization and radionuclide enrichment Earth Planetary Science Letters. 52, 345-354. Farai, I.P; R.I. Obed and N.N. Jibiri. (2006) Soil radioactivity and incidence of cancer in Nigeria. Journal of Environm ental Radioactivity. 90:1, 29-36. Forsberg, S., K. Rosn, V. Fernandez and H. Juhan. (2000) Migration of 137Cs and 90Sr in undisturbed soil profiles under controll ed and close-to-real conditions. Journal of Environmental Radioactivity. 50:3, 235-252. Fox, R.L., 1982. Some highly weathered soils of Puerto Rico, 3. Chemical properties. Geoderma. 27: 139--176. Heiri, O., Lotter, A.F. and Lemcke, G. ( 2001) Loss on ignition as a method for estimating organic and carbonate content in sediment s: reproducibility and comparability of results. Journal of Paleolimnology. 25, 101-110. Hunter, J. M. and S. I. Arbona. (1995) Paradise Lost: an introduction to the geography of water pollution in Puerto Rico. Soc. Sci. Med. 40 (10), 1331-1355. Jackson M.L. (1958) Soil Chemical Analysis -Advanced Course, published by the author, Madison, Wisconsin. Jibiri, N. (2001) Assessment of health risk levels associ ated with terrestrial gamma radiation dose rates in Nigeria. E nvironment International. 27:1, 21-26.

PAGE 152

139 Johnson-Pyrtle A. (1999) Distribution of 137Cs in the Lena River estuary-Laptev Sea system as evidenced by marine, estuarine and lacustrine sediments. PhD Thesis, Texas A&M University, College Station, TX. 206p. Johnson-Pyrtle, A., M. Scott, T. Laing a nd J. Smol. (2000) Cs-137 Distribution and Geochemistry of Lena River (Siberia) Drainage Basin Lake Sediments. The Science of the Total Environment. 255(1-3), 145-159. Johnson-Pyrtle, A. and M. Scott. (2001) Di stribution of Cs-137 in the Lena River Estuary-Laptev Sea System. Marine Pollution Bulletin. 42(10), 912-926. Infante R. and I. L. Acosta. (1991) Size dist ribution of trace metals in Ponce. Puerto Rico Air Particulate Matter. Atmo spheric Environment. 25B:1, 121-131. LaBrecque, J.J., P.A. Rosales and P.R. Co rdoves. (2001) Anomalously high values of cesium-137 in soils on the Peninsula de Paraguana (Venezuela). Journal of Radioanalytical and Nuclear Chemistry. 247 (3), 563-566. LaBrecque, J. J. and P. R. Cordoves. (2004) The spatial distribution of cesium-137 in surface soils in the Gran Sabana and Sie rra de Pacarima (Venezuela). Journal of Radioanalytical and Nuclear Chemistry. 262:2, 375-381. Larsen, M.C., Simon, A., 1993. Rainfall-thres hold conditions for landslides in a humidtropical system, Puerto Rico. Geography Annotaded. 75A 1–2., 13–23. Larsen, M.C., Parks, J.E., 1997. How wide is a road? The association of roads and masswasting disturbance in a forested montane environment. Earth Surface Processes Landforms. 229., 835–848. Larsen, M.C. and A. J. Torres-Sanchez. (1998) The frequency and distribution of recent landslides in three montane tropical re gions of Puerto Rico. Geomorphology. 24, 309–331. Larue, D.K. (1994) Puerto Rico and Virgin Island, Chapter 8. Caribbean Geology: An Introduction. U.W.I. Publis her Association, Kington. pp151-165

PAGE 153

140 Lindsay-Poland, J. (2001) Environmental im pacts of Navy training, Vieques Issue Brief [on-line] ms/PuertoRico/Envirobrief1101.html Livingston, H.D. & P.P. Povinec. (2000) Anth ropogenic marine radioactivity. Ocean and Coastal Management, 43, 689-712 Manahan, S., (2005) Environmental Chemistry 8th ed. CRC Press LLC. Printed in the United States. 783 pp. Micromeritics Instrument Corporation. 2003. Saturn DigiSizer 5200, Operator’s Manual. Missana, T., M. Garcia-Gutierrez, and U. Alonso. (2008) Sorption of strontium onto illite/smectite mixed clays. Physics a nd Chemistry of the Earth. 33, S156–S162. Moon, D.S., Hong, G.H., Kim Y.I.; Baskaran, M.; Chung, C.S., Kim, S., et al. (2003) Accumulation of anthropogenic and natural radionuclides in bottom sediments of the Northwestern Pacific Ocean. Deep Sea Research II. 50, 2649-2673. Montgomery, J.R. and R.J. Santiago. (1978) Zinc and Copper in ‘Particulate’ Forms and ‘Soluble’ Complexes with I norganic or Organic Ligands in the Guanajibo River and Coastal Zone, Puerto Rico. Estuarin e and Coastal Marine Science. 6, 111116. Morais Barros, A.J., S. Prasad, V. Duar te-Leite and A. G ouveia-Souza. (2007). Biosorption of heavy metals in upflow sludge columns. Bioresource Technology. 98, 1418–1425. Myasoedov, B. F. (2000) Century of radioche mistry: history and future. Journal of Nuclear and Radiochemical Sciences.1, 23-26. O'Brien, W. (1996) Bronze Age Copper Mining in Britain and Ireland. Shire Archaeology. Princes Risborough, 64pp.

PAGE 154

141 Odum H. (1970) Puerto Rico Nuclear Center Annual Report. The rainforest project, annual report, operated by the University of Puerto Rico under contract with the US Atomic Energy Commission, Rio Piedras, Puerto Rico. Odum, H. & G. Drewry. (1970) A tropical ra inforest; A study of i rradiation and Ecology at El Verde, Puerto Rico. Oa k Ridge, TN: Atomic Energy Commission Olsen, C.R, I.L. Larsen, N.H. Cutshall, J. F. Donoghue, O.P. Bricker and H.J. Simpson. (1981) Reactor-released radi onuclides in Susquehanna River sediments. Nature. 294, 242 245 Pett-Ridge, J. C., L. A. Derry and A.C. Kurtz. (2009) Sr isotopes as a tracer of weathering processes and dust inputs in tropical granitoid watershed, Luquillo Mountains, Puerto Rico. Geochi mica et Cosmochimica Acta. 73, 25–43. Poppe, L.J., V.F. Paskevich, J.C. Hathaw ay, and D.S. Blackw ood. (2001) A Laboratory Manual for X-Ray Powder Diffraction. U. S. Geological Survey Open-File Report 01-041. 88p Puerto Rico Nuclear Center. 1970. The rainforest project, annual report, operated by the University of Puerto Rico under c ontract with the US Atomic Energy Commission Puerto Rico Nuclear Center. 1969. The rainforest project, annual report, operated by the University of Puerto Rico under contract with the U.S. Atomic Energy Commission Puerto Rico Nuclear Center. 1968. The rainforest project, annual report, operated by the University of Puerto Rico under contract with the U.S. Atomic Energy Commission Renken, R., W.C. Ward, I.P. Gill, F. GomezGomez and J. Rodriguez-Martinez. (2000) Geology and Hydrogeology of the Caribbean Islands Aquifer System of the Commonwealth of Puerto Rico and the U. S. Virgin Islands. Professional Paper 1419, 148p.

PAGE 155

142 Robinson W. and V. E. Noshkin. (1999) Radi onuclide characterization and associated dose from long-live radionuclides in closein fallout delivered to the marine environment at Bikini and Enewetak Atolls. The Science of the Total Environment. 237/238, 311-327. Sadasivian, S, V.K. Shukka, S. Chinnaesa a nd S.J. Sartandel. (2003) Natural and fallout radioactivity measurements in Indian soils. Journal of Radioanalytical and Nuclear Chemistry. 256 (3), 603-607. Santchi, P.H. & B.D. Honeyman. (1989) Ra dionuclides in aquatic environment. Rad. Phys. Chem. 34(2), 213-240 Santschi, P., J.W. Murray, M. Baskaran, C.R. Benitez-Nelson, L.D. Guo, C.C. Hung, et al. (2006) Thorium speciation in seawater. Marine Chemistry. 100, 250–268. Scatena, F.N., and Larsen, M.C., 1991. Physi cal aspects of Hurricane Hugo in Puerto Rico. Biotropica. 23 4A, 317–323. Takeda, A., H. Tsukada, Y. Takaku, S. Hisa matsu, & M. Nanzyo (2006) Accumulation of uranium derived from long-term fertilizer applications in a cultivated Andisol. Science of The Total Environment. 367:2-3, 924-931. Tyler, A. N., S. Carter, D. A. Davidson, D. J. Long and R. Tipping. (2001) The extent and significance of bioturba tion on Cs-137 distributions in upland soils. Catena. 43, 81–99. United Nations Scientific Committee on the Effects of Atomic Radiation. (1982) Ionizing Radiation: Sources and Biological Effects Report. United Nations Scientific Committee on the E ffects of Atomic Radiation (UNSCEAR). (2000) The 2000 report to the General Assemb ly with scientific annexes. New York. United Nations Scientific Committee on the E ffects of Atomic Radi ation (2001) Report to the General Assembly.

PAGE 156

143 US Agency for Toxic Substances and Dis ease Registry. (2002) Dr inking water supplies and groundwater pathway evaluation, Isla de Vieques bombing range, Vieques, Puerto Rico. [on-line]: http://www. US Department of Agriculture, National Forest Service. (2004) History of the Caribbean National Forest. Retrieve on April 20, 2008. bean/about/history.shtml US Department of Homeland Security (1999) Surface Air Sampling Program. Retrieve on January 10,2008 US Department of Energy. (2003) DOE/EA-1394: Public access to the BONUS facility environmental assessment for authorizing the Puerto Rico Electric Power Authority (PREPA) to allow public a ccess to the boiling nuclear superheat (BONUS) reactor buildin g, Ricon, Puerto Rico [online].10-58 U.S. Department of Energy, Office of Legacy Management. (2005) Long-term surveillance and maintenance plan for th e Boiling Nuclear Superheated (BONUS) Reactor Facility, Rincon, Puerto Rico. Stoller Corporation, Colorado. US Environmental Protection Agency, 1996, Ur ban soil lead abatement demonstration project. Volume 1: EPA Integrated Re port, National Center for Environmental Assessment Office of Research and Deve lopment, US Environmental Protection Agency, Research Triangle Park, NC. US Environmental Protection Agency. (2001) Final Master Work Plan, Atlantic Fleet Weapons Training Facility. [On-line]: US Navy. (2002) Expenditure of depleted uranium (DU) rounds at Vieques Inner Training Range, Puerto Rico [on-line] US Natural Resources Conservation Services (2003) Soil Survey of Caribbean National Forest and Luquillo Experimental Forest ,Commonwealth of Puerto Rico. 182pp.

PAGE 157

144 US Nuclear Regulatory Commissi on, Office of Nuclear Material s Safety and Safeguards. 2000. Environmental Survey Report for Vieques Island Meetings and Press Conference. [On-line] http :// secys/2000/secy2000-0206/2000 0206scy.html#ENCLOSURE%20A US Nuclear Regulatory Commissi on, Office of Nuclear Material s Safety and Safeguards. (2001) Materials Licenses in 10 CFR Part 20 for the University of Puerto Rico, Ro Piedras Campus (NRC License 52-01986-04). Uyttenhove, J., M. Lemmens & M. Zizi. (2002) Depleted Uranium in Kosovo: results of a survey by gamma spectrometry on soil samples. Health Physics 83, 543-548. Valiela, I., and J.L. Bowen. (2002) Nitrogen sour ces to watersheds and estuaries: role of land cover mosaics and losses within watersheds. Environmental Pollution. 118, 239–248. Van Genderen E.J., A.C. Ryan, J. R. Tomass o, and S. J. Klaine (2005). Evaluation of acute copper toxicity to larval fathead minnows (Pime phales promelas) in soft surface waters. Environmental Toxico logy and Chemistry. 24:2, 408-414. Webb, Richard M.T. and Gmez-Gmez, Fe rnando. (1998) Synoptic Survey of Water Quality and Bottom Sediments, San Juan Bay Estuary System, Puerto Rico, December 1994-July 1995: U.S. Geological Survey Water Resources Investigations Report 97-4144. 69 p. World Health Organization. ( 2001) Depleted Uranium: S ources, Exposure and Health Effects. WHO/SDE/PHE/01.1. Geneva, April 2001, adiation/pub_meet/ir_pub/en/ Wyttenbach, A., V. Furrer and L. Tobler. (1995 ) The concentration ra tios plant to soil for the stable elements Cs, Rb and K. The Science of the Total Environment. 173/174, 361-367. Yarrow, M. (2000) History of the U.S. military in Puerto Rico: The U.S. Military in Latin America after the Cold War, Ameri can Friends Service Committee Latin

PAGE 158

145 American and Caribbean Peace Building Program. pp. 12 [on-line] rica/peace/strings/pricorep.pdf Zhiyanski, M., J. Bech, M. Sokolovska, E. Lucot, J. Bech and P.M. Badot. (2008) Cs137 distribution in forest floor and su rface soil layers from two mountainous regions in Bulgaria. Journal of Geochemical Exploration. 96:2-3, 256-266. Zikovsky, L. (2006) Alpha radioactivity in drin king water in Quebec, Canada. Journal of Environmental Radioactivity. 88:3, 306-309.

PAGE 159


PAGE 160

147 Appendix A. Gamma Spectrometry Calibration Energy (keV) 02004006008001000120014001600 Efficiency 0 10 20 30 40 50 60 70 Gamma calibration was performed using NIST 4357 Standard multiline. Energy (keV) Efficiency (%) Error ComputedError Difference 60.01 54.452.8854.692.89-0.24 86.54 61.573.0860.303.331.28 105.31 59.502.9860.483.13-0.97 122.06 58.202.9158.792.97-0.60 136.48 56.942.9756.523.000.42 391.70 23.112.2522.312.380.80 661.66 12.690.6312.720.64-0.03 834.83 9.840.4910.040.53-0.20 1115.55 7.490.387.320.410.17 1460.75

PAGE 161

148 Appendix B. X-Ray Diffraction Calibration Curves Calibration curve for XRD was performed using reference material from the Clay Mineralogical Society. Illite Calibration Kaolinite Calibration Curve

PAGE 162

149 Appendix C. Loss on Ignition Prot ocol (Heiri et al 2001; Dean, 1974) Step 1: Finely grind sediments and homogenize by either stirring by hand or shaking in a closed container. Step 2: Oven dry powdered samples in a pre-weighe d crucible (or container used for the heating) at 105 degrees Celsius fo r a time period between 1224 hours. Step 3: Cool to room temperature in desiccator (s) and weigh both samples and crucible to obtain the dry weight for the sample (DW 105) in grams. Step 4: Return sample to furnace and heat between 500-550 degrees Celsius for about 4hrs (when using mixed sediment samples). By heating the samples for 4hrs, the differences between weight losses between the samples are small enough that it may not be regarded as significant. Step 5: Cool to room temperature in desiccator (s) and weight sample in crucible to obtain the dry weight of sample (DW 550) in grams. Step 6: Calculate Loss on Igniti on using equation below. LOI550 = ((DW 105 DW 550) / DW 105) 100 Suggested sample weight of about 1-2 grams. However, when conducting this analysis it is important that all samples are approximately the same weight. This is to further reduce the difference between weight loss between the samples due to positional heating influence on the sample. Heated and cooled crucibles should be handled with thongs preferably. Sample spreadsheet setu p for recording data. Crucible weight (grams) Sample weight (grams) Dry Weight at 105 (grams) Dry Weight at 550 (grams) LOI Organic Matter (%) A sample ID column will be added to the actual table

PAGE 163

150 Appendix D. EPA Methods 3050b for ICP Analysis The recommended determinative technique s for each element are listed below: FLAA/ICP-AES Aluminum Magnesium Antimony Manganese Barium Molybdenum Beryllium Nickel Cadmium Potassium Calcium Silver Chromium Sodium Cobalt Thallium Copper Vanadium Iron Zinc Lead Vanadium This method is not a total di gestion technique for most sa mples. It is a very strong acid digestion that will dissolve almost all el ements that could become “environmentally available.” By design, elements bound in silica te structures are not normally dissolved by this procedure as they are not usually mobile in the environment. If absolute total digestion is requi red use Method 3052. Summary of Method For the digestion of samples, a represen tative 1-2 grams (wet weight) or 1 gram (dry weight) sample is digested with repeated additions of nitric acid (HNO3) and hydrogen peroxide 3 (H2O2). For ICP-AES or FLAA analyses, hydrochloric acid (HCl) is added to the initial digestate and the sample is refluxed. In an optio nal step to increase the solubility of some metals (see Section 7.3.1: NOTE), this digest ate is filtered and the filter paper and residues are rinsed, first with hot HCl and then hot reagent wa ter. Filter paper and residue are returned to the digestion flask, refluxed with additional HCl and then filtered again. The digestate is then diluted to a final volume of 100 mL. If required, a separate sample aliquot sha ll be oven-dried for a total percent solid determination. Apparatus and Materials Digestion Vessels 250-mL Vapor recovery device (e.g., ribbed wa tch glasses, appropriate refluxing device, appropriate solvent handling system) Drying ovens able to maintain 30EC + 4EC. Temperature measurement device capable of measuring to at least 125EC with suitable precision and accu racy (e.g., thermometer, IR sensor, thermocouple, thermister, etc.) Filter paper Whatman No. 41 or equivalent. Centrifuge and centrifuge tubes Analytical balance capable of accurate weighing to 0.01 g.

PAGE 164

151 Heating source Adjustable and able to maintain a temperature of 90-95EC. (e.g., hotplate, block digesters, microwave, etc.) Funnel or equivalent. Graduated cylinder or equivale nt volume measuring device. Volumetric Flasks 100-mL. Reagents Needed Reagent grade chemicals shall be used in all tests. Reagent Water Nitric acid (concentrated), HNO3 Hydrochloric acid (concentrated), HCl Hydrogen peroxide (30%), H2O2 CAUTION: Wear goggles and gloves at all ti mes!!!! Lab coat is recommended. Procedure: 1. Mix sample to homogeneity and sieve using a USS Standard #10 (2mm) sieve. For each sample, weight to the nearest 0.01 g (1-2 g of wet sample and 1 g of dry sample). 2. Add 10 mL 1:1 HNO3 (1 part DI water to 1 part conc. HNO3), mix and cover with a watch glass or vapor r ecovery device. 3. Reflux (stir using glass rod or ma gnetic stirrers). Heat at 95oC 5o without boiling for 10 to 15 minutes, and allow sample to cool. 4. Add 5 mL conc. HNO3. 5. Replace the cover and reflux for 30 minutes at 95oC 5o. 6. If brown fumes are generated (indi cating oxidation of the sample by HNO3) repeat this step by adding 5 mL of conc. HNO3. Repeat as needed until NO brown fumes are given off by the sample (i ndicating a complete reaction with HNO3). 7. Return sample to hotplate and allow sa mple to evaporate to approximately 5 mL by heating at 95oC max, without boiling, for two hours. Maintain a covering of solution over the bottom of the vessel at all times. DO NOT allow complete evaporation or drying. 8. Allow sample to cool to room temper ature. Add 2 mL of water and 3 mL of 30% H2O2. Cover with watch glass and warm to about 60oC with hotplate. 9. CAREFULLY continue to add 1mL aliquots of 30% H2O2 until bubbling subsides. DO NOT add more than 10 mL of 30% H2O2 TOTAL. 10. Cover the sample with a ribbed watch glass and continue heating (max of 95oC) until the volume has been reduced to approximately 5 mL.

PAGE 165

152 11. Add 10 mL Conc. HCl to the sample. 12. Cover and reflux (stir using glass rod or magnetic stirrers ) for 15 minutes. 13. Filter the sample through a pre-weighed and pre-labeled Whatman No. 41 filter paper and collect filtrate liquid in 100 mL volumetric flask. 14. Dry filter paper. After drying weigh filter paper plus re sidue subtracting original weight of pape r to find residue weight. Steps 15-26 may be used to improve the so lubilities and recoveries of Antimony (Sb), Barium (Ba), Lead (Pb), an d silver (As) when necessary. *Heat 1L of hot water to 95oC and sit aside. 15. Add 2.5 mL conc. HNO3 and10 mL conc. HCl to 1-2 g wet weight or 1g dry weight) sample and cover with watch glass. 16. Place sample on heat source at 95oC, reflux (stir using glass rod or magnetic stirrers) for 15 minutes. 17. Filter the digestate through Whatman No. 41 filter paper and collect filtrate in 100 mL volumetric flask. 18. Wash filter paper while still in the funnel with no more than 5 mL hot (95oC) HCl, then with 20 mL hot (95oC) DI water. 19. Collect washings in the same 100 mL flask containing previous filtrate. 20. Remove filter paper and residue from the funnel, and place both the filter paper w/ residue back in to the vessel (beaker). 21. Add 5 mL conc. HCl and heat at 95oC until filter paper dissolves. 22. Remove vessel from heat source and wash the cover and sides with DI water. 23. Filter the residue and collect the filtrat e in same 100 mL volumetric flask used for previous filtrate. 24. Allow filtrate to cool, then dilute to volume. If preci pitate occurs in the flask upon cooling, DO NOT dilute to volume. 25. If precipitate forms add up to 10m L HCl to dissolve the precipitate. 26. Dilute to volume needed for analysis. (100 ml)

PAGE 166

153 Appendix E. Sample preparation for XRD Analysis Removal of CaCO3 & Mg CO3 with NaOAc buffer 1. Amount of sediment start with depe nds on composition of sample (mostly sand or clay) = 20g sample needed 2. Separate sample into amounts of 5g & put each in a 100mL centrifuge tube (if highly calcareous 10-20% CaCO 3, or sandy soil & shell, treat in beaker instead of centrifuge tubes. 3. Add 50mL of NaOAc bufferpH5 to each 5g sample of soil. 4. Stir with rubber tipped rod. 5. Digest in near boiling bath for 30 minutes with intermittent stirring. 6. Centrifuge until supernatant liquid is cl ear, liquid is decanted & discarded. 7. 2 more washing with 1N NaOAc are done & centrifuge if sample is nearly calcareous, repeat water bath also Removal of Organic Matter MnO2 by H2O2 1. With sample from above (that’s st ill wet with NaOAc buffer) transfer sediment from centrifuge bottle to beaker (600mL) using a rubber policeman. Try to use as little wa ter in transferring as possible. 2. Add 5mL 30% H2O2. 3. Stir & cover with watchglass. Watch carefully to prevent bubbling over. If bubbling over, squirt with DI water or stir in cold water bath to control reaction. 4. Once reaction has subsided, place on hotplate for few minutes & watch closely. 5. Stir & rinse sides of be aker with 5mL of H2O2. 6. After reaction is stable add 10mL more of H2O2 & cover glass. Let digest on hotplate for 2-4 hours covered. 7. Repeat 6 as needed until complete dige stion of O.M. is complete (black to grayish white). 8. When the dark color of O.M. has largely disappeared, wash soil 3 times with NaOAc centrifuge & decant. Wash once with 95% methanol, centrifuge & decant. Removal of free Ferric Iron 1. Using sample from before (5-10g) place in centrifuge 100mL centrifuge bottle. 2. Add 40mL of 0.3 M Na-citrate & 5mL of 1 M NaHCO3 3. Place in hat water bath at Between 75-80 degrees Celsius (not more) 4. Add 1 g of solid Na2S2O4 using spoon & stir for 1 minute then occasionally for 5 minutes. 5. Add 2nd amount of 1g Na2S2O4 & stir 6. Add 3rd amount of 1g Na2S2O4 at end of 2nd 5 minute period. 7. Heat aver burner for 15 minutes between 75-80 degrees Celsius. 8. Add 10mL of NaCl & 10mL of methanol. 9. Mix, warm, & centrifuge for 5 minutes at 1600-2200rpm

PAGE 167

154 10. Decant & save supernatant in 500mL volumetric flask. 11. If brown or red color persists, repeat treatment. Then treat with 1N NaCl, centrifuge & decant. Add 10mL methanol, mix & warm in water bath (do not let boiling occur). Ce ntrifuge for 5 minutes.

PAGE 168

155 Appendix F. Data from adsorption/desorption experiments Metal desorption on clay minerals. Concentration express in nM. Illite pH Cu Rb Cd Cs Pb 6.05 33.87 2494.51 23.40 249.88 0.06 6.18 28.29 2006.60 0.33 198.17 0.36 6.20 32.31 2331.87 31.63 233.25 0.02 6.22 77.22 2630.23 -0.08 245.59 0.75 6.84 140.26 2427.81 0.24 223.24 0.11 6.85 63.53 2368.90 0.28 201.05 0.30 Kaolin pH Cu Rb Cd Cs Pb 4.44 1.12 0.31 0.33 -0.27 4.76 1.78 1.28 0.25 0.48 -0.54 5.33 1.11 1.01 -0.17 0.20 -1.03 5.6 1.12 1.24 0.25 0.42 -1.40 5.94 1.16 1.26 -0.16 0.40 -0.73 6.35 1.53 1.55 -0.05 0.71 -0.22 6.78 1.52 1.25 -0.80 0.40 -0.54 8.53 1.06 1.22 -0.74 0.40 -0.72 Montmorillonite pH Cu Rb Cd Cs Pb 5.21 24.17 151.29 0.65 6.01 0.44 5.44 19.31 151.64 1.70 6.07 0.08 5.47 24.38 216.10 1.98 10.62 0.62 6.33 17.39 88.88 0.31 3.44 0.04 6.93 19.72 195.75 0.66 8.19 0.14 7.36 32.87 140.40 0.34 5.88 0.08

PAGE 169

156 Metal desorption by site, concentration express in nM. Espiritu Santo Estuary pH Cu Rb Cd Cs Pb 6.31 226.92 260.80 20.87 18.37 0.56 6.36 101.14 133.73 1.77 0.42 0.40 6.44 101.23 146.02 1.83 0.32 8.41 6.56 237.62 342.12 2.50 0.65 1.34 6.56 163.50 274.84 8.24 10.43 0.72 6.65 225.19 203.12 2.26 0.35 0.66 6.81 659.36 166.26 4.95 0.52 0.38 7.12 363.36 136.78 2.99 0.28 0.29 7.2 204.73 90.83 2.14 0.31 8.31 El Verde Experimental Station pH Cu Rb Cd Cs Pb 4.4 337.08 121.33 4.92 2.68 7.91 4.4 227.87 46.97 5.41 4.58 1.72 4.6 285.15 99.98 3.08 2.47 0.65 4.6 249.74 51.79 3.84 3.81 0.57 4.89 180.34 28.51 1.40 2.51 0.81 5.24 307.34 81.81 1.02 1.82 0.43 5.24 201.43 28.00 1.41 2.43 1.28 5.64 294.75 73.29 0.40 1.42 0.35 5.81 237.78 24.89 0.13 1.78 0.09 5.9 337.08 121.33 4.92 2.68 7.91 BONUS area, Rincon pH Cu Rb Cd Cs Pb 6.56 93.30 341.53 1.72 0.95 0.03 6.62 57.89 242.43 2.29 1.14 0.57 6.63 90.03 303.04 1.43 0.89 0.06 6.66 78.93 269.67 1.81 1.21 0.38 6.77 58.24 305.61 2.07 1.70 0.26 6.82 95.27 282.09 0.04 0.99 0.13 6.86 97.90 290.99 0.66 0.53 7.90 7.01 105.12 271.21 -0.08 0.81 0.72 7.14 79.78 126.93 0.55 0.77 7.29 7.43 93.30 341.53 1.72 0.95 0.03

PAGE 170

157 Vieques pH Cu Rb Cd Cs Pb 6.8 93.62 323.98 6.92 1.51 0.64 6.83 248.01 500.66 2.79 1.54 0.72 6.91 255.09 432.56 2.49 1.53 0.41 6.92 88.96 183.34 0.83 0.88 0.55 6.94 90.17 309.47 7.48 1.62 0.20 7.12 62.82 90.08 2.37 0.80 8.36 7.14 308.91 325.62 1.16 1.45 0.86 7.15 40.00 130.69 3.60 0.85 0.24 7.3 119.60 185.03 1.43 0.86 8.05 7.58 93.62 323.98 6.92 1.51 0.64

PAGE 171

158 Distribution coefficient for e quilibrium kinetics experiments on clay minerals, express in nM. Illite Time Cu Rb Cd Cs Pb 0 1482.55 1102.99837.02718.78 460.28 10 77.44 1325.65429.23431.14 12.10 30 95.88 1271.82400.32406.38 10.31 60 77.85 1419.25405.65432.11 9.03 90 54.45 1412.23370.34411.65 7.80 120 1446.16353.35390.35 8.01 180 73.05 1519.87328.08385.46 6.98 360 78.57 1561.99299.04362.44 6.16 Montmorillonite Time Cu Rb Cd Cs Pb 0 1401.66 1155.41841.95711.56 441.22 10 16.87 373.24118.2773.76 5.45 30 12.09 352.6593.4162.50 3.71 60 12.27 357.9188.9063.50 5.49 90 13.46 357.5683.8862.76 3.44 120 17.28 354.0582.7962.28 2.87 180 17.83 353.9382.3260.63 3.08 360 10.13 360.7281.5562.60 3.65 Kaolin Time Cu Rb Cd Cs Pb 0 1495.45 1131.42848.72714.87 455.69 10 417.49 1010.91650.51563.11 12.93 30 372.64 1024.95627.30559.87 7.31 60 334.88 1032.32618.49573.19 7.04 90 310.48 1033.02611.19561.75 6.39 120 291.91 1028.93607.73563.41 5.81 180 270.51 1021.91601.45561.91 5.79 360 217.64 1031.85598.78557.32 5.20

PAGE 172

159 Distribution coefficient for adsorption experi ments on clay minerals, express as the log Ks. Illite pH Cu Rb Cd Cs Pb 5.94 1.50 -1.16 0.71 0.33 2.12 6.29 1.57 -0.46 1.05 0.47 2.32 6.34 1.54 -0.78 0.85 0.43 2.82 6.84 1.67 -0.62 1.00 0.42 3.00 7 1.54 -0.30 1.13 0.47 3.06 7.01 1.37 0.02 1.42 0.55 2.90 Montmorillonite pH Cu Rb Cd Cs Pb 4.79 1.54 0.73 1.20 1.55 2.83 5.13 1.88 0.54 0.99 1.30 2.49 5.46 1.99 0.82 1.26 1.56 2.73 6.14 1.39 0.90 1.42 1.63 2.88 6.88 1.41 0.53 1.43 1.36 2.01 7.31 1.80 0.85 1.69 1.59 3.27 Kaolin pH Cu Rb Cd Cs Pb 4.86 2.04 -0.21 0.32 0.24 3.30 5.59 1.53 -0.21 0.20 0.26 3.16 5.94 1.74 -0.22 0.27 0.28 3.05 6.06 1.76 -0.11 0.59 0.31 3.05 6.4 1.40 -0.43 1.05 0.12 1.81 6.89 1.52 -0.23 1.04 0.27 2.24 8.34 2.04 -0.21 0.32 0.24 3.30

PAGE 173

160 Distribution coefficient for adsorption experi ments at the study sites in Puerto Rico, express as the log Ks. Espiritu Santo pH Pb Cu Cd Cs Rb 6.30 3.38 0.85 1.94 1.55 0.13 6.38 2.65 0.45 1.88 1.58 0.23 6.43 2.11 0.77 1.59 1.56 -0.02 6.46 2.12 0.75 1.66 1.58 -0.01 6.57 2.93 0.55 1.87 1.60 0.17 6.64 2.92 0.82 1.61 1.59 0.04 6.66 3.38 0.85 1.94 1.55 0.13 6.81 2.65 0.45 1.88 1.58 0.23 7.12 2.11 0.77 1.59 1.56 -0.02 El Verde pH Pb Cu Cd Cs Rb 4.36 1.83 0.50 0.58 0.98 0.25 4.37 2.03 0.68 0.54 0.85 0.29 4.49 2.06 0.68 0.74 1.06 0.36 4.66 2.75 0.77 0.81 1.00 0.46 4.72 2.20 0.51 0.33 0.80 0.25 5.05 2.91 0.86 1.25 1.11 0.63 5.34 2.05 0.61 1.39 1.05 0.40 5.52 3.00 0.72 1.82 1.28 0.77 5.65 2.86 0.62 1.75 1.20 0.49 5.81 2.81 0.57 1.88 1.34 0.61 Rincon pH Pb Cu Cd Cs Rb 6.57 3.77 1.22 1.82 1.28 0.08 6.62 3.43 1.16 2.00 1.39 0.13 6.66 3.01 1.24 1.66 0.95 -0.03 6.72 2.18 0.73 1.99 1.34 0.09 6.80 2.99 1.29 1.71 0.88 -0.04 6.82 2.00 1.28 1.85 1.01 -0.02 6.88 2.28 1.19 2.10 1.35 0.14 7.04 2.91 1.08 2.19 1.42 0.21 7.11 3.60 1.21 1.74 0.45 -0.67 7.22 2.22 1.08 1.90 0.47 -0.48

PAGE 174

161 Vieques pH Pb Cu Cd Cs Rb 6.85 3.23 1.06 1.64 1.08 -0.13 6.89 2.17 0.81 2.58 1.12 -0.39 6.89 2.82 0.81 2.55 1.32 -0.10 6.89 3.56 0.70 1.97 1.37 -0.06 7.00 2.70 1.11 1.68 1.02 -0.19 7.12 2.17 0.71 2.07 1.23 -0.36 7.19 2.19 1.15 1.68 0.94 -0.42 7.24 2.18 1.20 1.83 0.85 -0.32 7.43 3.34 0.86 2.36 1.27 -0.08 7.60 4.09 1.24 1.97 0.88 -0.20

PAGE 175

162 Appendix G. Self-adsorption and weight corrections In order to verify the radionuclide activi ties determined during this investigation, a subset of samples were analyzed uti lizing two gamma detectors with different configurations. Activity values reported in chapter 2 were obtained utilizing a gamma detector with a well configuration. Both the gamma well and planar detectors were calibrated using the same calibra tion standards. In addition, self-adsorption curves were calculated as a function of the weight of each standard. Determining the self-adsorption characteris tics and efficiencies of both detectors allowed for the verification of Cs-137 ac tivity values determined during this investigation. Environmental samples tend to ha ve low levels of radioactivity. As such it is important to be able to accurately determine the efficiency of each detector, particularly at 100 keV and below for this study. Figure 53 and 54, shows the selfadsorption trend and weight correction factor s for samples analyzed using the planar detector. The full energy peak efficiencies are affected by the high self-absorption of the gamma rays emitted, which strictly depend on the energy of the gamma-ray considered as well as on the composition and apparent density of the analyzed sample. As the total weight of the sample increase self-adsorp tion chances will increase. Figure 53and 54 shows that correlation and how using samples in a range of weights that were from 1gram to 11grams, will allow us to correct for the tr ansmission received by planar detector as a function of sediment weight.

PAGE 176

163 Figure 53. Pb-210 self adsorption correction on planar detector, R squared is equal to 0.99. Figure 54. Pb-210 weight vs. efficiencies on planar det ector. Pb-210 transmission received by planar detector as a function of sediment weight. Well detector efficiency is plotted in figur e 55. Efficiency of the well detector is greater at lower energy levels. 0 5 10 15 20 25 0.005.0010.0015.001/Eff. (decays/count)Weight (g) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.005.0010.0015.00([Pb-210 counts]/[Pb-210 (o)counts])Weight (g)

PAGE 177

164 Energy (keV) 02004006008001000120014001600 Efficiency 0 10 20 30 40 50 60 70 Figure 55. Energy Efficiency calibra tion for Germanium well detectors Self adsorption directly affects the coun ting of samples since the adsorption is a function of sample density (Cochran et al. 1998). Therefore environmental samples need to be normalized in order to eliminate varia tions in density which affects transmission to the detector (Cochran et al 1998). This normalization is performed using the same geometry, realizing that alt hough the geometry and volume of a sample can be controlled, the density of a sample varies accordi ng to the composition of the sample. Environmental samples are a mix of organic ma tter, different soil types, mineral content, as well as various other parameters (includi ng moisture content) that can easily change during dry seasons, storms or other weather-re lated events. Because the island of Puerto Rico has a tropical climate, storms, hurricanes, wave action, etc. can directly impact the chemical, biological and geological com position of environmental samples.

PAGE 178

165 Samples from Vieques were measured for Pb-210 activity utilizing both the gamma well and planar detect ors (figure 56). Activity values determined varies throughout the core, and decrea ses from zero to 16 cm and increases again in deeper sediments. The consistency between the two de tectors is strong and se rves to validate the acceptable usage of both detectors during this investigation.

PAGE 179

166 Planar detector Pb-210 Activity (dpm/g) (cm) 0 5 10 15 20 25 30 Pb-210 Activity (dpm/g) analyzed on well detector 0 5 10 15 20 25 30 Figure 56. Result comparison of Pb-210 act ivity in samples from Mosquito Bay performed on well and planar detector.

PAGE 180

ABOUT THE AUTHOR Warner Ithier-Guzman was born in Mayaguez, Puerto Rico. He has a B.S. Degree in Biology from the Inter-American Univers ity-San German, a M.S. Environmental Protection and Evaluation from Inter-Ameri can University-Metropolitan Campus. As a doctoral student he conducted research on radiogeochemistry and work at the Environmental and Occupational Safety and H ealth Office at the University of Puerto Rico-Rio Piedras. He received the NSF Bridge to the Doctorate Fellowship as well as the OCEANS Fellowship. Some of the specialized course work on his curriculum includes special seminars from the Vespucci Institute and the Radiation Safety Officer Certification from Harvard University. He has been appointed to various committees including the Advancing Science in Li mnology and Oceanography local organizing committee held in San Juan, Puerto Rico in February 2011.