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 Ka
controlfield tag 001 002001024
007 cr mnu|||uuuuu
008 090427s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002540
Dietz, Marianne E.
Investigating environmental change due to hypoxic conditions on the Louisiana continental shelf :
b a geochemical approach
h [electronic resource] /
by Marianne E. Dietz.
[Tampa, Fla] :
University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 95 pages.
Thesis (M.S.)--University of South Florida, 2008.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
ABSTRACT: The Louisiana (LA) shelf is chronically affected by seasonal hypoxia that has been shown to be spatially expanding and growing progressively more severe. Hypoxic conditions on the shelf have been closely linked to the large quantities of nutrients delivered to the Gulf of Mexico via the Mississippi River. Multiple geochemical proxies on a suite of sediment cores from the LA shelf provide a record of environmental change that parallels the onset of hypoxic conditions over the last century and prior to anthropogenic influences. The sedimentary record for the last century shows a shift from terrestrial to algal sources to the sediments, as well as a trend of increasingly enriched nitrogen isotopes, which is probably caused by a combination of large amounts of denitrification, either within the Mississippi watershed or on the continental shelf, increased primary productivity, compounded with relatively minor increases of enriched nitrogen source inputs. The current chronic hypoxia, especially since the early 1970's, is exacerbated by anthropogenic nutrient loading from the Mississippi River basin, but other processes must be responsible in the past. In the historic record, several episodic low-oxygen events are defined by forminiferal assemblages and shifts in the geochemical records. Geochemistry of three sediment cores from the Louisiana (LA) shelf indicates that these historic events are likely caused by increased inputs of terrestrially-derived organic matter during peaks in Mississippi River discharge. These results suggest that low-oxygen conditions may be a natural, aperiodic phenomenon on the LA shelf, and that the current seasonally severe hypoxia resulted from the excess anthropogenic nutrient input.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: David J. Hollander, Ph.D.
x Marine Science
t USF Electronic Theses and Dissertations.
Investigating Environmental Change Due To Hypoxic C onditions On The Louisiana Continental Shelf: A Geochemical Approach by Marianne E. Dietz A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science College of Marine Science University of South Florida Major Professor: David J. Hollander, Ph.D. Edward S. VanVleet, Ph.D Nancy N. Rabalais, Ph.D Date of Approval: July 3, 2008 Keywords: hypoxia, isotopes, sediment, nitrogen, hi story Copyright 2008, Marianne E. Dietz
i Table of Contents List of Tables ii List of Figures iii Abstract iv Chapter 1: Introduction and Background 1 Introductory Statements 1 Global Coastal Hypoxia 3 Eutrophication Process 5 Hypoxia on the Louisiana Shelf 7 The Mississippi River Basin 12 Hypoxia: A Recent Phenomenon? 16 Study Objectives 18 Scientific Approach 19 Chapter 2: Methods 29 Core Selection 29 Total Organic Carbon 33 Stable Isotope Analyses 34 Chapter 3: A Geochemical Perspective of Hypoxic Co nditions on the Louisiana Continental Shelf Over the Past Century 37 Abstract 37 Introduction 38 Methods 40 Results 44 Discussion 48 Summary 57 Chapter 4: A Multi-Proxy Sedimentary Record of His torical Low-Oxygen Conditions on the Louisiana Continental Shelf 59 Introduction 59 Methods 61 Results 64 Discussion 68 Chapter Five: Summary 74 References Cited 77 Appendix A: Geochemistry Data 87
ii List of Tables Table 1 Modeled total nitrate load estimates from t he Mississippi River basin from two different studies 11 Table 2 Land use in the Mississippi River basin 14 Table 3 Core information, including locations, wate r depth, length, estimated accumulation rate, and frequency of mid-summer hypo xia 31 Table 4 Information on the three sediment cores use d in this study 41
iii List of Figures Figure 1. Locations of major hypoxic zones around t he world 4 Figure 2. Schematic of the eutrophication process o n a continental shelf 6 Figure 3. Estimated nitrogen fertilizer use in the United States (green line) and nitrate concentrations in the Mississippi River from 1955-1995 (blue line) 10 Figure 4. Nitrogen inputs to the Mississippi River Basin 12 Figure 5. Mississippi River drainage basin and subbasins 13 Figure 6. Changing spatial extent of the hypoxic zo ne on the LA shelf as a results of annual monitoring 16 Figure 7. Distinctive source values of atomic C/N a nd 13C for algae and land plants 27 Figure 8. Map showing locations of the three sedime nt cores used in this study 30 Figure 9. PEB index data for the three cores in thi s study 32 Figure 10. Map showing core locations for cores use d in this study 40 Figure 11. PEB index data for the three cores in th is study 43 Figure 12. Geochemistry (TOC, 13C, C:N) results from each of the three cores 47 Figure 13. Stable nitrogen isotopes and percent nit rogen results for the three cores 50 Figure 14. Map showing core locations relative to f requency of hypoxic conditions on the LA shelf 62 Figure 15. Geochemistry results for PE0305-GC1 65 Figure 16. Suggested time windows based on geochemi cal results. 70
iv Investigating Environmental Change Due to Hypoxic C onditions on the Louisiana Continental Shelf: A Geochemical Approach Marianne E. Dietz ABSTRACT The Louisiana (LA) shelf is chronically affected by seasonal hypoxia that has been shown to be spatially expanding and growin g progressively more severe. Hypoxic conditions on the shelf have been c losely linked to the large quantities of nutrients delivered to the Gulf of Me xico via the Mississippi River. Multiple geochemical proxies on a suite of sediment cores from the LA shelf provide a record of environmental change that paral lels the onset of hypoxic conditions over the last century and prior to anthr opogenic influences. The sedimentary record for the last century shows a shi ft from terrestrial to algal sources to the sediments, as well as a trend of inc reasingly enriched nitrogen isotopes, which is probably caused by a combination of large amounts of denitrification, either within the Mississippi wate rshed or on the continental shelf, increased primary productivity, compounded with rel atively minor increases of enriched nitrogen source inputs. The current chroni c hypoxia, especially since the early 1970Â’s, is exacerbated by anthropogenic n utrient loading from the Mississippi River basin, but other processes must b e responsible in the past. In the historic record, several episodic low-oxygen ev ents are defined by
v forminiferal assemblages and shifts in the geochemi cal records. Geochemistry of three sediment cores from the Louisiana (LA) shelf indicates that these historic events are likely caused by increased inputs of ter restrially-derived organic matter during peaks in Mississippi River discharge. These results suggest that low-oxygen conditions may be a natural, aperiodic p henomenon on the LA shelf, and that the current seasonally severe hypoxia resu lted from the excess anthropogenic nutrient input.
1 Chapter 1: Introduction and Background Introductory Statements The onset and expansion of the hypoxic zone on the Louisiana (LA) shelf has been largely attributed to anthropogenic influe nces, particularly an increase in the use of nitrogen-based commercial fertilizers within the Mississippi River drainage basin beginning in the 1950Â’s (Turner & Ra balais 1991, Goolsby & Battaglin 2001, Goolsby et al. 2001, Turner & Rabal ais 2003). Based on field observations conducted since the 1985, both the sev erity of oxygen depletion and the spatial extent of the chronically hypoxic z one on the Louisiana continental shelf have increased significantly (Rab alais et al. 1994, Rabalais et al. 2001, Rabalais et al. 2007a, Rabalais et al. 20 07b). The subsequent increase in organic carbon production induces high amounts o f respiration in the water column and sediments, which leads to oxygen depleti on in bottom waters. The low-oxygen bottom waters on the LA shelf are largel y unsuitable to sustain many types of marine organisms, threatening the local ec osystem and therefore endangering the lucrative fishing industry in the G ulf of Mexico (Rabalais & Turner 2001a, Adams et al. 2004). Measurements of hypoxic conditions were first recor ded in the early 1970Â’s, but measurements prior to then do not exist Several paleoindicators in accumulated sediments (as reviewed by Rabalais et a l. 2007b) show a
2 worsening of hypoxia in conjunction with anthropoge nic nutrient input. These shifts in paleoindicators are consistent with seaso nally severe hypoxia beginning in the middle of the 20th century as anthropogenic reactive nitrogen compoun ds began to increase (Galloway & Cowling 2002). Prior to anthropogenic hypoxia, paleoindicators in older sediments indicate aperiod ic but similar foraminiferal peaks in lower oxygen conditions (Osterman et al. 2 005). This suggests that low oxygen events may have occurred prior to the define d period of anthropogenic hypoxia. Determining the natural physical and biolo gical processes along with anthropogenic factors in the formation of low oxyge n conditions on the LA shelf is crucial to understanding the dynamics of hypoxia an d determining management needs. Geochemical analyses of the sedimentary record can be used to extend the historic record beyond the instrumental record and effectively evaluate environmental changes. Bulk sedimentary analyses c an provide organic matter source information as well as record paleoenvironme ntal information (Meyers 1997). Here, a multi-proxy approach to analyzing se diment cores from the Louisiana continental shelf is used to reconstruct environmental conditions associated with the modern anthropogenic hypoxic an d historical low-oxygen events. A combination of bulk organic geochemical s edimentary analyses, such as total organic carbon (TOC), stable isotopic rati os of carbon ( 13C) and nitrogen ( 15N), and C:N ratios, provide an assessment of the ch anging sources of organic matter and insight into the environmenta l response to the onset,
3 expansion and increasing severity of recent hypoxia and conditions during past low-oxygen events. This study investigates the geochemical parameters of three cores from the LA shelf that span periods of current anthropog enic hypoxia and prior low oxygen events and indicators of environmental condi tions in the past. Results of this study will supplement the paleoindicator resul ts of other researchers with additional geochemical approaches. This thesis is divided into an introduction, a met hods section, a summary chapter, and two chapters that have been structured as manuscripts for peerreviewed journals. As such, some material and figur es are repetitive. Global Coastal Hypoxia Coastal hypoxia is a growing environmental problem around the world. As the worldÂ’s population increases, particularly in c oastal regions, increasing stress is placed on coastal ecosystems. Hypoxia in marine systems is typically defined as dissolved oxygen concentrations less than 2 mg O2 l-1, while anoxia is the complete lack of dissolved oxygen in the system (Ni xon 1995). Hypoxia has been found to occur naturally in some coastal zones and parts of the ocean, and has existed on geologic timescales. Well known areas of human-caused or humanaggravated hypoxia around the world include the Che sapeake Bay, parts of the Black, Baltic, and Adriatic Seas, Long Island Sound and the northern Gulf of Mexico (Diaz 2001). In these areas, respiration exc eeds oxygen resupply through limited surface to bottom exchange of oxygen (Figur e 1).
4 Figure 1 : Locations of major hypoxic zones around the world (figure from http://earthtrends.wri.org) The occurrence of hypoxic and anoxic environments i n coastal areas worldwide appears to be increasing (Diaz & Rosenber g 1995, Cloern 2001, Diaz 2001). The rise in the number of coastal zones affe cted appears to be largely due to anthropogenic activities, as many of these h ypoxic zones have not historically been hypoxic. Early European settlemen t of the Chesapeake Bay area changed the watershed dramatically and promote d the eutrophication of the Bay, which has grown more severe with time (Zimmerm an & Canuel 2000, 2002). Similar trends are seen in the northern Adri atic Sea, where increasing human populations correspond with decreasing oxygen concentrations (Justi et al. 1987, Justi 1991). Increasing human population is closely coup led to
5 increased reactive nitrogen from fossil fuel burnin g, fertilizers, and leguminous crops (Galloway & Cowling 2002). Anthropogenic flux es of nitrogen and phosphorus, in particular, are now significantly hi gher than historically (Nixon 1995, Cloern 2001). Hypoxic conditions can pose a great threat to marin e organisms and the health of marine ecosystems (Diaz & Rosenberg 1995, Rabalais & Turner 2001a). When dissolved oxygen concentrations drop b elow critical limits, significant impacts start to occur to marine life a nd ecosystems. These conditions threaten water quality and can impair marine ecosys tems, a valuable food source and economic industry. While some marine organisms can adapt to lower oxygen concentrations, others may become stressed to the p oint of mortality. This results in reduced biodiversity, mortality of benthic commu nities, loss of biomass, stresses on fisheries, and alteration of ecosystem trophic structure. Eutrophication Process Hypoxia occurs as a result of the eutrophication pr ocess. Eutrophication is the increased rate of primary production and carbon accumulation in the environment (Nixon 1995). The organic matter produc ed through primary production in the surface waters sinks and is decom posed either in the water column or bottom sediments via the respiration proc ess, which consumes available oxygen. If the oxygen utilized is not rep laced, dissolved oxygen concentrations in the water drop and hypoxic condit ions develop (Figure 2). Eutrophication can cause noxious and toxic algal bl ooms, decreased light
6 penetration, oxygen depletion, and subsequent loss of habitat and biodiversity (Killops & Killops 2005). Although eutrophication c an be a natural phenomenon, it often is a result of increased nutrient loading in marine systems. Anthropogenic changes in the proportion of these nutrients can al ter or exacerbate hypoxic conditions in coastal waters (Nixon 1995, Cloern 20 01, Rabalais et al. 2004). Figure 2: Schematic of the eutrophication process on a conti nental shelf (figure from www.conservationinstitute.org) Stratification is also a necessary requirement for hypoxia to develop. Water column stratification effectively restricts v ertical mixing and therefore the replenishment of the water column oxygen supply fro m the atmosphere to the bottom waters. Stratification is a density differen ce between water masses. Wind stress can also play a factor in whether hypoxic co nditions will develop. Strong episodic winds can reoxygenate bottom waters by mix ing a stratified water
7 column, but this usually only lasts for short perio ds of time (Rabalais et al. 1994, Wiseman Jr. et al. 1997). Conversely, low wind stre ss allows stratification to develop and persist, which is favorable for the dev elopment of hypoxic condtions (Walker & Rabalais 2006). Hypoxia in the Northern Gulf of Mexico/Louisiana Sh elf The Louisiana (LA) continental shelf in the norther n Gulf of Mexico is one of the largest human-caused coastal hypoxic areas i n the world, and is the largest area of coastal hypoxia in the United State s. Like many other hypoxic zones around the world, this region is chronically affected during the spring and summer months by seasonal hypoxic conditions, when dissolved oxygen concentrations drop below 2 mg O2 L-1 in the water column (Rabalais & Turner 2001a). Hypoxia arises through stratification via s ummer outflow of Mississippi River water and the delivery of nutrients which cau se algal blooms. Low-oxygen bottom water conditions are most commonly found on the shelf in water depths of 5-30 meters, but have been documented in water dept hs of up to 60 meters (Rabalais et al. 1998, Rabalais et al. 1999, Rabala is et al. 2002c, Rabalais et al. 2007b). Hypoxic conditions can occur from the sedim ents up through most of the water column, but usually only encompasses the lowe r portion (20-50%) of the water column depending on the total water depth and the depth of the pycnocline (Rabalais & Turner 2001b, Rabalais et al. 2002c). The LA shelf is dominated by the outflow of the Mis sissippi River, which delivers large amounts of both nutrients and fresh water to the shelf. The
8 development of hypoxia west of the Mississippi delt a is dependent on several interrelated physical and biological factors. Stron g vertical stratification and sufficient amounts of organic matter inputs are req uired to deplete bottom water oxygen concentrations. Stratification on the LA she lf generally increases in the spring when freshwater discharge from the continent increases (Rabalais & Turner 2001b, Walker & Rabalais 2006) which results in severe and widespread hypoxia on the shelf (Rabalais et al. 2007b). A rel atively warm freshwater lens comprised of Mississippi River water overlays the c ooler, denser water of the Gulf, creating a strong pycnocline that can be up t o 10 meters thick and can effectively isolate bottom waters to replenishment from oxygen-rich surface waters (Rabalais et al. 2002b). Stratification can persist longer into the fall depending on the timing of the breakdown of stratif ication by winds or storms (Rabalais & Turner 2001b, Rabalais et al. 2002c). D uring the winter, strong storms cause mixing and allow the bottom waters to be oxygenated frequently, and therefore hypoxia seldom forms. Every year, large quantities of nutrients from the continent are delivered via the Mississippi and Atchafalaya Rivers to the L A shelf, causing large annual algal blooms. The observed changes in nutrient conc entrations have resulted in increased primary production in the northern Gulf o f Mexico (Justi et al. 1993, Rabalais et al. 1996, Scavia et al. 2003). Supporti ng evidence for this theory includes measurements of increased chlorophyll conc entrations and enhanced primary productivity in the Mississippi River plume (Lohrenz et al. 1990, Lohrenz et al. 1997), long term patterns in the accumulatio n of Â“biologically boundÂ” silica
9 (Turner & Rabalais 1994), and seasonal-scale cohere nce between river discharge and net productivity estimated from oxyge n time-series (Justi et al. 1993). There appears to be a direct correlation between ri verborne nutrient inputs and primary production on the continental shelf wit hin the Mississippi RiverÂ’s influence. Highest values of primary production are observed in the spring, while the lowest values are observed in the fall (Lohrenz et al. 1997). Additionally, primary productivity on the continental shelf appea rs to be moderate when not enhanced by the high Â“newÂ” nitrogen nutrient loads that enter via riverine outflows. Since freshwater discharge varies markedl y on seasonal and interannual time scales, it has been suggested that Â“discharge drivenÂ” is a more appropriate description of the primary productivity of this subtropical continental margin than is its spatial partition into regions b ased on mean productivity (Biggs & Sanchez 1997). Anthropogenic activities have increased the amount of nutrients delivered to the Gulf of Mexico. Nutrient loading through use of anthropogenic agricultural fertilizers is thought to be a leading cause of a 3 00% increase in nitrogen loads to the LA shelf (Goolsby et al. 1999, Donner et al. 20 02, Justi et al. 2003). A majority of all commercial fertilizers used in the United States are applied to cropland within the Mississippi River basin (Figure 3) (Goolsby et al. 1999). Previous studies have shown that nitrate concentrat ions in the river have doubled over the past 35 years (Turner & Rabalais 1991, Bra tkovich et al. 1994).
10 Figure 3: Estimated nitrogen fertilizer use in the United St ates (green line) and nitrate concentrations in the Mississippi River from 1955-1995 (blue line) (Goolsby et al. 1999) The delivery of such large amounts of freshwater, s ediment and nutrients through the Mississippi River plume dramatically af fects the biological and biogeochemical processes on the continental margin. Results from multiple studies have shown that nutrients delivered by the Mississippi River are a dominant factor controlling hypoxia on the Louisian a shelf (Rabalais et al. 1996, Lohrenz et al. 1997, Rabalais et al. 1999, Scavia e t al. 2003, Turner et al. 2006).
11 Source (%) Alexander et al. 2008 Booth & Campbell 2007 Fertilizers 67 59 Atmosphere 16 17 Manure 5 13 Sewage 9 11 Other 3 N/A Table 1: Modeled total nitrate load estimates from the Missi ssippi River basin from two different studies Many studies have concluded that fertilizer runoff from agriculture is the main source of nitrogen for streams and rivers in t he Mississippi River basin (Figure 4 and Table 1) (Goolsby et al. 1999, Alexan der et al. 2000, Goolsby & Battaglin 2001, McIsaac et al. 2001, Donner 2003, D onner & Kucharik 2003, Donner et al. 2004, Alexander et al. 2008). Large s cale production of commercial nitrogen-based fertilizers only became possible aft er the development of the Haber-Bosch process for the synthesis of NH3 using atmospheric nitrogen (Nixon 1995). Intensive fertilizer use has implications fo r coastal eutrophication not just because greater application will result in greater runoff, but because the amount of fertilizer lost from fields may increase with in tensity of application (Nixon 1995).
12 Figure 4: Nitrogen inputs to the Mississippi River Basin (fi gure from USGS) Residual nutrients may also play a role in the deve lopment of hypoxia. Surface water nitrate flux in the Mississippi River basin was found to be related to both the nitrogen inputs of the current year and a small contribution of the inputs from up to the previous 9 years (McIsaac et al. 200 1). Even organic matter produced in previous years and deposited in surface sediments may aid in the development of hypoxic conditions by increasing res pirations rates on top of respiration of Â“freshÂ” organic matter (Turner et al 2008). The Mississippi River Basin The Mississippi River is the major source of fresh water and sediments to the northern Gulf of Mexico. The Mississippi River is one of the largest rivers in the world, and drains the third largest basin in th e world, about 3.3 x 106 km2,
13 which comprises approximately 41% of the contiguous United States (Figure 5). Additionally, about 30% of Mississippi River water is now diverted through the Atchafalaya River. The drainage basin includes all or part of 31 different states and is one of the most productive farming regions i n the world. About 58% of the Mississippi River drainage basin is farmland and is one of the most productive farming regions in the world (Table 2). A large portion of the intensively used farmland i n the Midwest is normally too wet to farm. To circumvent this problem, farmer s have developed tile draining, where perforated tiles or pipes are used to contour the fields and move water into drainage ditches. This type of drainage system can greatly increase nitrate runoff from agricultural fields (McIsaac & Hu 2004), and a significant percentage of farm land in the Midwest contain tile -drains (David et al. 2006). Figure 5 : Mississippi River drainage basin and sub-basins ( figure from USGS).
14 Federal farm policies and commodity payments that s upport fertilizer and chemical-intensive cropping practices may be some o f the most important factors that shape the nutrient-related consequences of lan d use in the Mississippi River basin. These subsidies encourage the expansion of f ertilizer-intensive crops, and decrease land diversification. Although these farm programs protect farmers from economic pitfall and uncertainties, they have led t o farming practices that are not environmentally friendly (Booth & Campbell 2007), n or economically stable over the long term. Land Use % Cropland 58% Woodland 18% Range and Barren land 21% Wetlands and water 2.4% Urban 0.6% Table 2: Land Use in the Mississippi River basin (CENR 20 00) In addition to a steady population increase within the Mississippi basin with related inputs of nitrogen through municipal w astewater systems, human activities have changed the natural functioning of the Mississippi River system. Human development has dramatically altered the rive r basin. Approximately 70 million people live within the basin (Goolsby et al 2001). The suspended
15 sediment load of the Mississippi has been changed d ramatically due to the addition of dams and levees along the river and its tributaries during the last century, which has decreased by approximately half since the 1800Â’s, most of that since 1950 (Corbett et al. 2004). Deforestatio n, agricultural practices, and changes in land management strategies have also con tributed to changing the sediment load of the Mississippi River (McKee & Bas karan 1999, Turner & Rabalais 2003). Dams on the Missouri and Arkansas R ivers have reduced the amount of sediment that makes it to the Mississippi River. Conversely, poor land practices on the Ohio River have increased the amou nt of sediment that is delivered to the Mississippi (Corbett et al. 2007). As population continues to increase and food production increases, it is likel y that land use will continue to change (Table 2) and nutrient use will increase as well. Mississippi River discharge varies seasonally due t o climatic factors. The flow maximum takes place in April, the minimum in S eptember, and shows a strong seasonal pattern in nutrient inputs. Flow te nds to peak in spring or early summer, when runoff from precipitation events and d rainage from tile systems is common (Battaglin et al. 2001). Grown crops tend to limit runoff from fields and tile drains later in the year. Maximum sediment dis charge coincides with maximal flow (in the spring). The sediment load in the spri ng is approximately five times greater than in the fall, when flow is at a minimum (Corbett et al. 2004). Sediment delivery to the Gulf of Mexico is heavily influence d by sediment storage and remobilization in the lower Mississippi. When river discharge drops, sediment is temporarily stored in the lower river until dischar ge remobilizes the sediment and
16 delivers it to the Gulf of Mexico (McKee & Baskaran 1999) Storage periods are typically 4-8 months, depending on yearly river flo w patterns. Figure 6: Changing spatial extent of the hypoxic zone on the LA shelf as a result of annual monitoring (figure from N. Rabalai s, LUMCON) Hypoxia: A Recent Phenomenon? Systematic annual monitoring of the hypoxic zone be ginning in 1985 has documented a significant increase in the extent and severity of hypoxic conditions on the LA shelf during the past few deca des (Figure 6) (Rabalais et al. 1994, Rabalais et al. 1999, Rabalais et al. 2001, R abalais et al. 2007b). Additionally, proxy studies also indicate an overal l increase in continental shelf oxygen stress in intensity, duration and spatial ex tent in the last 100 years, particularly since the 1950s (Turner & Rabalais 199 1, Eadie et al. 1994, Rabalais
17 et al. 1999). Previously published studies using pa leoindicators have identified several pre-anthropogenic low-oxygen events, as wel l as hypoxic conditions that have gradually become more severe over the past cen tury. Accumulations of the mineral glauconite were found to be particularly hi gh in areas of documented hypoxia on the LA shelf, and show increases in sedi ments coincident with increased use of anthropogenic fertilizers (Nelsen et al. 1994). This trend has also been documented by analysis of bacterial chlor opigments in dated sediment cores from the Mississippi River bight (Chen et al. 2001). Increases in the amount of biogenic silica in sediments also suggest a gradual worsening of lowoxygen conditions over long time scales and particu larly within the last few decades (Turner & Rabalais 1994). Additionally, sev eral benthic foraminiferal proxies have been developed to reconstruct changes in oxygen conditions in the past. Faunal assemblages from surface and sediment cores on the LA shelf show that low-oxygen tolerant species of foraminife ral increase with time, further suggesting that hypoxic conditions have worsened (B lackwelder et al. 1996, Platon et al. 2005). The A-E index, which looks at the percentages of two benthic species relative to one another, also indicates pro gressively worsening lowoxygen condition on the LA shelf (Sen Gupta et al. 1996). The onset and expansion of the hypoxic zone is larg ely attributed to anthropogenic influences, particularly an increase in the use of nitrogen-based commercial fertilizers within the Mississippi River drainage basin beginning in the 1950Â’s (Goolsby & Battaglin 2001, Goolsby et al. 20 01). Based on field observations conducted since 1985, both the severit y of oxygen depletion and
18 the spatial extent of the chronically hypoxic zone on the LA continental shelf have increased significantly (Figure 6) (Rabalais et al. 1994, Rabalais et al. 2001, Rabalais et al. 2007b). The subsequent increase in organic carbon production induces high amounts of respiration in the water co lumn and sediments, which leads to oxygen depletion in bottom waters (Dortch et al. 1994, Turner et al. 1998, Rowe et al. 2002, Quinones-Rivera et al. 2007 ). The low-oxygen bottom waters on the LA shelf are largely unsuitable to su stain many types of marine organisms, threatening the local ecosystem and ther efore endangering the lucrative fishing industry in the Gulf of Mexico (R abalais & Turner 2001a, Adams et al. 2004). Study Objectives This study investigates the environmental effects o f hypoxic conditions on the LA shelf during recent and historic time period s. Analyses of several sediment cores provide a detailed record of environ mental history and changing conditions in the local ecosystem. Results of this study will be used to expand the instrument record prior to the 1970Â’s and provide a comprehensive historical record of low-oxygen conditions on the LA shelf. Th e specific objectives and research questions of this project are as follows: 1. Investigate the amount organic matter loading on th e LA shelf a. Has there been a change in the amount of organic ma tter in LA shelf sediments in relation to increased nutrient i nput and severity of oxygen depletion?
19 2. Evaluate the source of organic matter to the sedime nts a. What is the origin of the organic matter in LA shel f sediments? b. Have these sources changed over time? 3. Examine nutrient sources and cycling in the marine environment on the LA shelf a. Has there been a change in nutrient source or cycli ng evident in the sedimentary record that corresponds to hypoxic conditions? 4. Compare and contrast modern hypoxic conditions to h istoric lowoxygen events Scientific Approach Geochemical analyses of the sedimentary record can be used to extend the historic record beyond the instrumental record and supplement existing paleoindicator studies. Bulk sedimentary analyses can provide organic matter source information as well as record paleoenvironme ntal information (Meyers 1997). Here, a multi-proxy approach to analyzing se diment cores from the Louisiana continental shelf is used to identify mod ern anthropogenic hypoxic and historical low-oxygen events and correlate them wit h known geochemical parameters that indicate environmental change. A co mbination of bulk organic geochemical sedimentary analyses, such as total org anic carbon (TOC), stable isotopic ratios of carbon ( 13C) and nitrogen ( 15N), and C:N ratios, provide an assessment of the changing sources of organic matte r and insight into the
20 environmental response to the onset and increasing severity of recent hypoxia and past low-oxygen events. Total Organic Carbon The TOC in sediments is simply a measure of the cha nging amount of organic material in the sediments. Environmental co nditions influence how much of the originally produced organic matter is incorp orated into the sedimentary record (Meyers 1997). Under hypoxic conditions, an increase in the TOC in the sediments may be a measure of carbon preservation ( Demaison & Moore 1980). Alternatively, some researchers would argue that in creased productivity, not an increase in organic carbon preservation, would resu lt in an increase in sedimentary TOC (Calvert & Pedersen 1992). Although the Mississippi River delivers some organic matter to the Gulf Of Mexico, most of the organic material that reaches the sediments is from in situ phytoplankton production (Eadie et al. 1994, Turner & Rabalais 1994, Justi et al. 1997, Rabalais & Turner 2001b). Addtionally, it is not clear what happens to the or ganic matter produced in this region, and some organic matter may be transported elsewhere on the shelf or off the shelf entirely (Lohrenz et al. 1997). If th e organic matter is incorporated into shelf sediments, it would result in a greater chance of developing hypoxic conditions in benthic environments (Justi et al. 1993). A possible explanation for the ultimate fate of algal-derived OM is that it si nks near the plume and is transported along the shelf by westward flowing bot tom currents (Wiseman Jr. et al. 1997).
21 Stable Isotopes Stable isotopes are an effective tool in environmen tal studies and can act as natural tracers of geochemical and biological pr ocesses (Fry 2007). Due to slight mass differences, the stable isotopes of an element are very similar but not identical. These differences lead to isotopic fract ionation during chemical, physical and biological processes which can be meas ured using mass spectrometry (Burdige 2006). In ecosystem studies, stable isotopes can provide information about energy flows, nutrient sources, a nd trophic relationships between producers and consumers (Peterson & Fry 198 7, Fogel & Cifuentes 1993). Stable isotopes can be used to infer the bio logical source of organic material, the ecological conditions in which that o rganic material was formed, and the fate of that organic matter (Fogel & Cifuentes 1993). Carbon Stable Isotopes There are two stable isotopes of carbon that occur in nature: 12C (98.90%) and 13C (1.10%). The stable isotopic ratios of carbon ( 13C) in bulk sediments give a first-order indication of the major source o f carbon to the sediments (Hedges & Parker 1976, Meyers 1994, O'Reilly & Heck y 2002). Carbon isotope ratios are typically reported in standard per mil n otation in reference to the PDB standard using the following equation:
22 Specific enzymes are responsible for carbon isotope fractionation in photosynthetic plants. These enzymes differ between plants that use the C3 pathway for photosynthesis and plants that follow t he C4 pathway. C3 plants, which are most of the deciduous, broadleaf plants, utilize the rubisco enzyme which has a fractionation of about -20Â‰. C4 plants, or plants that prefer drier climates like many types of grasses, use the enzyme PEP carboxylase, which as a fractionation around -7Â‰ (Meyers 1994). Terrestri al plant organic material is usually centered around -27Â‰ for C3 vascular plants or -16Â‰ for C4 plants. Algal organic material averages about -22Â‰ (Killops & Kil lops 2005, Pancost & Pagani 2006). The Mississippi River drainage basin is comprised m ostly of C3 plants which tend to have 13C values around -27Â‰ (Sackett 1964, Hedges & Parker 1976). As such, the 13C values of terrestrially-derived organic carbon fr om the Mississippi River Basin are distinctly different fr om the carbon isotope values from algal organic matter, which are typically cent ered around -21Â‰ near the mouth of the Mississippi River (Sackett 1964). This offset between algal and terrestrial plant organic material is useful for de lineating between the relative contribution of algal and terrestrial organic matte r sources to the sediments (Sackett 1964, Eadie et al. 1994, Meyers 1994). Add itionally, it has been suggested that erosion of C4 plant soils or substantial sediment inputs from
23 western Mississippi River tributaries could also sk ew the carbon isotopic values towards more marine values of -22Â‰ (Wissel & Fry 20 05). Carbon isotopic values from sediments on the LA she lf tend to reflect mostly marine organic matter input. Surficial sedim ent TOC 13C values in the northern GOM vary in one study from -21.7 to -19.7Â‰ (Goni et al. 1997). Autochthonous marine organic matter appears to domi nate over terrestriallyderived organic matter in the Gulf of Mexico, espec ially in non-coastal areas (Hedges & Parker 1976). Nitrogen Stable Isotopes There are two naturally occurring nitrogen stable i sotopes: 14N (99.63%) and 15N (0.36%). Stable isotopes of nitrogen ( 15N) are can be used to discriminate between trophic levels in ecosystems ( O'Reilly & Hecky 2002), to evaluate the source and cycling of nitrate in sedim ents (Macko et al. 1993, Kendall 1998, Chang et al. 2002, Panno et al. 2006) trace anthropogenic pollution inputs (Fry 2007) or changes in environme ntal conditions over time (Meyers 1997). Nitrogen isotope ratios are typicall y reported in standard per mil notation relative to an atmospheric nitrogen standa rd using the following equation:
24 Biological processes that incorporate nitrogen-cont aining compounds into organisms discriminates between the heavy and light isotopes of nitrogen, tending to favor the incorporation of 14N over 15N (Battaglin et al. 2001). In exclusively marine environments, if sources of nitr ogen from land can be eliminated, nitrogen stable isotopes can be used to trace the biological mechanisms in which phytoplankton and bacterial inc orporate nitrogen into their biomass (Macko et al. 1993). However, eliminating land-based sources of nitrogen is difficult to do, especially in coastal zones. Different sources of nitrogen can have distinctly different isotopic rations. Nitrate that originates from sewage and/or livestoc k has different 15N values (~8 20Â‰) than nitrate from soil (~2 15Â‰), atmospher ic deposition (~0Â‰) or anthropogenic fertilizer (-4 +2Â‰). It should be noted that there is significant overlap observed among some sources, and that there is some inherent uncertainty associated with the nitrogen stable iso tope proxy (Kohl et al. 1971). Marine organic matter, which in coastal waters is p roduced using land-based nitrogen sources, tends to be much more depleted th an its source pool unless enough productivity occurs to completely exhaust th e nutrients available (Macko et al. 1993). Lake studies have shown that fraction ation during uptake by phytoplankton can be on the order of -4 to -5Â‰ (Fog el & Cifuentes 1993). Wastewater from treated sewage and manure is also t ypically enriched in 15N. Natural microbial processes in tertiary sewage t reatment plants are strongly discriminatory fractionation processes that selecti vely utilize 14N and produce a 15N-enriched wastewater signature. Treatment plant ef fluent values have even
25 been reported to be greater than 30Â‰ (Savage & Elmg ren 2004). This heavy nitrogen signal can then be traced through coastal areas as it is transported physically and biologically through ecosystems. Mea suring stable isotopic ratios in biota and sediments can provide a useful indicat or for tracing spatial and temporal impacts of sewage-derived nitrogen in aqua tic ecosystems (Savage 2005). Sedimentary 15N records can also reflect bacterial diagenetic pro cesses. Like other biological processes, bacterial processe s preferentially utilize and remove the lighter isotope of nitrogen ( 14N) first, leaving behind the heavier isotope ( 15N) (Berner 1971, Altabet 2006). This removal of the lighter isotope causes the 15N record to become more enriched downcore as more and more of the light isotope is removed and utilized. Stable isotopes of nitrogen can also be used as an indicator of the extent of denitrification. Denitrification is the biologic ally mediated process by which fixed nitrogen, specifically NO3, is reduced to N2. This process occurs in a wide variety of environments wherever oxygen supplies ar e limited and there are sufficiently abundant amounts of nitrate/nitrite an d organic matter. However, ecosystems do not have to be completely anoxic for denitrification to take place (Seitzinger et al. 2006). Denitrifying microbes pre ferentially utilizes nitrate with the 14N isotope over the 15N isotope and can have large fractionation ranges (Mehnert et al. 2007). Significant positive shifts in nitrogen isotope ratios can be a strong indicator that the denitrification process is consuming NO3 (Kellman & Hillaire-Marcel 1998).
26 Continental shelves are the regions where the large st amount of denitrification occurs (44% of the global total) (S eitzinger et al. 2006). On seasonally hypoxic continental shelves, like the LA shelf, nitrate from in situ production and external inputs is distributed throu ghout the water column by vertical mixing and is available for denitrificatio n when the waters become stratified and the bottom layer becomes hypoxic (Se itzinger et al. 2006). Water samples from the Mississippi River and tile drains within the Mississippi basin have been found to have 15N values between 4.8 and 16.4Â‰ (Panno et al. 2006). Additionally, samples of Mississippi River w ater flowing into the Gulf of Mexico have particulate organic matter with isotopi c ratios of nitrogen between 5-9Â‰ (Battaglin et al. 2001, Kendall et al. 2001). The large range of isotopic values found in the Mississippi River basin and riv er outflow on the shelf suggest that there are several processes that act on nitrog en isotopic ratios even before nutrient source pools reach the continental shelf. Carbon:Nitrogen Ratios Molar ratios of carbon to nitrogen (C:N) are also u seful indicators of organic matter source. As with stable carbon isotop e ratios, algal-and terrestrially-derived organic matter have distinct signatures, which can be detected in the sediment record (Jasper & Gagosian 1990, Meyers 1994). Algal and bacterial organic matter typically has C:N rati os between 4-10 (Wissel & Fry 2005), while terrestrial plant material usually has C:N ratios greater than 20 (Figure 7) (Ertle & Hedges 1985, Meyers 1994). The distinction between
27 terrestrial and marine values has to do with the am ount of cellulose within cells: algal cells lack the compound, and terrestrial plan ts contain abundant cellulose. Figure 7: Distinctive source values of atomic C/N and 13C for algae and land plants (Figure from Meyers 1994). Although C:N ratios can be subject to alteration du ring diagenesis in the water column and sediments, the original ratio is u sually fairly well-preserved in the sediment record (Meyers 1997). Variations in th e sedimentary record can also be used to examine diagenetic processes in sed iments. Increases in C:N usually is indicative of the preferential loss of n itrogen as organic matter is consumed during diagenesis, while similar C:N value s downcore suggest that carbon and nitrogen are being preserved or minerali zed in the same manner (Macko et al. 1993 and sources therein). Utilizing the combination of stable
28 isotope and C:N ratio proxies allows for clearer de termination of organic matter source than using only one proxy (Meyers 1994). C:N ratios within the Mississippi River are fairly stable around 9.9 +/0.3, indicative of primarily algal and/or bacterial sour ces (Trefry et al. 1994).
29 Chapter 2: Methods Study Site and Core Selection This study used three sediment cores located on the LA shelf on the periphery of current-day hypoxia (Figure 8). The lo ngest core (PE0305-GC1, 163 cm in length) represents a historical time-frame, w hile a shorter core (PE0305BC1, 31 cm in length) captures the past 100 years. Both of these core sites were located in 47 meters of water and were selected bec ause they are located in an area that experiences 25 to 50% frequency of occurr ence of mid-summer hypoxia (Rabalais et al. 2002b). An additional sedi ment core (MRJ05-BC6, 42 cm in length) also represents the most recent centu ry and was selected because it is located in a region of the LA shelf where the frequency of occurrence in midsummer is less than 25%. This core was located in 6 5 meters of water, approximately twenty meters deeper water than the o ther two cores, and was originally meant to represent an unaffected area. A ll of these cores were located in water depths that were deeper than where hypoxia is most commonly documented on the LA shelf (Rabalais et al. 1998, R abalais et al. 1999, Rabalais et al. 2002b).
30 Figure 8: Map showing locations of the three cores used in th is study relative to the frequency of mid-summer hypoxic conditions o n the LA shelf for the period 1985-2002 (from Rabalais et al. 2002 b) These cores were obtained and sampled every centime ter by scientists at the USGS (Table 3). 210Pb chronology on each core was also completed by USGS personnel and has been previously published (O sterman et al. 2005, Osterman et al. 2008). 210Pb is a natural component (t1/2 = 22.3) of atmospheric deposition that is strongly adsorbed to particles. These particles accumulate in sediments and can then be measured in sediments to estimate age to within approximately the past 100 years (Bierman et al. 19 98).
31 Core ID PE0305-GC1 PE0305-BC1 MRD05-BC6 Core Type Gravity Box Box Date Collected 7/28/02 7/28/02 5/12/05 Latitude 28 23.796 28 23.975 28.3970 Longitude 90 27.701 90 27.170 90.7091 Depth (mwd) 47 47 65 Length (cm) 164 31 42 Frequency of the Occurrence of Hypoxia# 25 to 50% 25 to 50% Less than 25% 210Pb rate (cm/yr) 0.33* 0.34* 0.36+ *published in Osterman et al. (2005), +Osterman et al. (2008), and #Rabalais et al. (2002a) Table 3 : Core information, including locations, water dep th, length, estimated accumulation rate, and frequency of mid-s ummer hypoxia A low-oxygen tolerant faunal assemblage, known as t he PEB index, has been found to be a faunal proxy for the occurrence of low-oxygen conditions on the LA shelf (Osterman et al. 2000, Osterman 2003). This proxy relies on the relative abundance of a few opportunistic foraminif eral species ( Pseudononion atlanticum, Epistominella vitrea and Buliminella morgani ). As benthic oxygen
32 concentrations drop, the species of foraminifers th at are low-oxygen intolerant will decrease in number or become locally extinct. Therefore, the abundance of the opportunistic species that can tolerate short i ntervals of low-oxygen conditions will become relatively higher than durin g normoxic conditions. It is important to note that this proxy cannot determine the precise value of oxygen concentration, but provides an indication of relati vely low-oxygen conditions. Figure 9: PEB index data for the three cores in this study (O sterman et al. 2005, Osterman et al. 2008). Note that the core fro m deeper water, MRD05-BC6, does show an increase in the PEB index, suggesting that this core also experiences hypoxic conditions, but to a lesser extent than the other two cores. The raw data are i ncluded in Appendix A.
33 PEB index data from previously published studies on the same cores used in this project indicate that all three cores are a ffected by periodic low-oxygen conditions in the past few decades (Osterman 2003, Osterman et al. 2005, Osterman et al. 2008, Swarzenski et al. 2008). The core from deeper water (MRJ05-BC6), where hypoxia is currently less freque nt than 25% in mid-summer, was previously thought to be unaffected by hypoxic conditions but appears to exhibit some changes in low-oxygen conditions. The PEB index in the two shallower cores, PE0305-GC1 and PE0305-BC1, increas e at approximately the same rate, while MRD05-BC6 begins to increase earli er (1930) than the other two and at a lower rate. PE0305-GC1 has the highest values in the most recent decades but does not begin to increase until about 1960. The other hypoxic core, PE0305-BC1 has values that are almost as high, but begins to increase just before 1940. Total Organic Carbon Sediments were dried in a drying oven at 60C for a t least 24 hours. The drying times were adjusted depending on the amount of water per sample. Concentrations of total organic carbon in each samp le were determined using a UIC Carbon Coulometer at the University of South Fl orida. Determination of the amount of organic carbon via coulometry involves me asuring the amount of carbon dioxide (CO2) evolved from sediment by converting the total car bon and inorganic carbon phases. Standards of calcium carbo nate were used during each run to determine instrument accuracy and to establi sh a standard response
34 curve. Each standard was weighed and analyzed on th e coulometer to establish a standard curve, which was used in the quantificat ion of each unknown sample. Analysis of total carbon involves the combustion of the sample to convert both organic and inorganic forms of carbon into CO2. A small amount of sample is weighed and placed in a porcelain boat. The samp le is then introduced into the carbon coulometer oven, where it is combusted at 97 0C in the presence of excess oxygen to ensure complete oxidation of the s ample into CO2. Inorganic carbon quantification is done by the addition of an acid to the sample. Another aliquot of sample was weighed and placed into a sma ll glass sample flask. The sample is then acidified in a heated reaction vesse l using 5mL of 2 M perchloric acid, which reacts only with the inorganic carbon a nd creates CO2. The CO2 generated by the acidification of the sample is mea sured and yields the amount of inorganic carbon. Once the total amount of carbo n per sample and the total amount of inorganic carbon per sample has been quan tified, the amount of organic carbon can be calculated by subtracting the amount of inorganic carbon from the total amount of carbon in each sample. Stable Isotope Analysis Bulk sediment isotope analyses ( 13C and 15N) were run using a Carlo Erba 2500 Series I elemental analyzer coupled with a continuous-flow Finnigan Mat Delta Plus XL stable-isotope mass spectrometer at the University of South Florida. Each sample was acidified with 0.1N HCl in a glass beaker to remove all inorganic carbonates. The sample was stirred and le ft in the acid overnight to
35 ensure that all carbonate had reacted with the acid After 24 hours, the acid was decanted, the sediment was rinsed with Milli-Q wate r and then decanted again. This process was repeated. The samples were then dr ied for 24 hours in a drying oven at 60C. When dry, the sediment was removed fr om the beaker and ground as finely as possible with a mortar and pestle to h omogenize the sample as much as possible. Each sample was weighed and packed inside a tin ca psule. A relatively large amount of sediment was needed per sample and was based on the amount of total carbon and total organic carbon in each sa mple to ensure a large enough (and therefore clear) signal response. Helium is us ed as a carrier gas which speeds up sample processing time, but inherently di lutes the sample, and is why a large sample size is needed. Additionally, each s ample was run in duplicate After being introduced into the elemental analyzer Â– isotope ratio mass spectrometer (EA-IRMS) system via autosampler, the sample is completely combusted at 1000C in the presence of oxygen. The combustion products are then reduced from NOx to N2. The resulting gas is passed via a helium carrier gas through a chromatographic column to separate di screte Â“packetsÂ” of carbon and nitrogen, and then goes through a thermoconduct ivity detector which causes the formation of an electrical pulse directly propo rtional to its quantity. This is how C:N ratios and %N values are calculated by computer software. After leaving the EA, the sample in gaseous form tr avels into the continuous-flow mass spectrometer by helium carrier gas. The CO2 and N2 enter into an ion source where they are ionized (stripped of an electron) and then
36 propelled through a magnetic field. All ions have t he same charge but different masses which affects their flight path through the magnetic field in that each ion is deflected through the magnetic field by an amou nt proportional to its mass. The ions are then collected by a detector, which me asures the two stable isotopes of carbon (12C and 13C) and nitrogen (14N and 15N) and allows the computer to calculate the ratio between the isotope s. This ratio is analyzed relative to a reference gas (CO2 for carbon and N2 for nitrogen) to report isotopic values in standard per mil notation relative to PDB for carbon isotopes and air international standards for nitrogen. Instrumental accuracy was monitored through analysi s of known standards at regular intervals. Spinach leaves were used as a standard. Average standard deviations of isotopic measurements for th e standards were 0.17Â‰ for nitrogen and 0.21Â‰ for carbon in this data set. Bla nks were used at the beginning of each run to correct for instrument err or.
37 Chapter 3: A Geochemical Perspective of Hypoxic Con ditions on the Louisiana Continental Shelf Over the Past Centu ry Abstract The Louisiana (LA) shelf is chronically affected b y seasonal hypoxia that has been shown to be spatially expanding and growin g progressively more severe (Rabalais et al. 1994, Rabalais et al. 2007b ). Hypoxic conditions on the shelf have been closely linked to the large quantit ies of nutrients delivered to the Gulf of Mexico via the Mississippi River. Multiple geochemical proxies on three sediment cores from the LA shelf provide a record o f environmental change that parallels the onset of hypoxic conditions over the last century. Carbon isotopic results suggest that the cores most affected by hyp oxic conditions have undergone a shift in organic matter source from ter restrially-derived to mostly in situ phytoplankton-derived. A core from deeper water th at is not as affected by low-oxygen does not show this shift, and has remain ed algally-dominated throughout the length of the core. In contrast, sta ble nitrogen isotopes in all cores show an increasingly enriched trend. Nitrogen isoto pic values prior to the use of anthropogenic fertilizers in the Mississippi River Basin are about 3Â‰, and increase to about 6Â‰ in recent years. The observed 15N values could be caused by denitrification in the Mississippi River basin or on the shelf, a change in the source of nitrogen from terrestrial to algal an increase in primary production, or a combination of factors. Increasing nitrogen flux to the Gulf of
38 Mexico corresponds temporally with increases in sed imentary total nitrogen on the LA shelf. Since it has been reasonably proven t hat in situ primary production has increased with corresponding increases of river -borne nutrients, it is most likely that an increase in primary production is at least partially responsible for the increasing 15N values in the sediments. The geochemical indices in this study indicate that this trend is probably caused by a co mbination of denitrification and increased primary productivity. Introduction The Louisiana continental shelf is chronically affe cted by seasonal hypoxia during spring and summer. Based on field observatio ns conducted since 1985, both the severity of oxygen depletion and the spati al extent of the chronically hypoxic zone on the LA continental shelf have incre ased significantly (Rabalais et al. 2001, Rabalais et al. 2004, Rabalais et al. 2007a, Rabalais et al. 2007b). Every year, large quantities of nutrients from the continent are delivered via the Mississippi River to the Gulf of Mexico and cause l arge annual phytoplankton blooms and increases in primary production (Lohrenz et al. 1990, Dagg & Breed 2003). The subsequent increase in organic carbon pr oduction induces high amounts of respiration in the water column and sedi ments, which, coupled with strong water column stratification, leads to oxygen depletion in bottom waters (Rabalais et al. 2002a, Rabalais et al. 2007b). The low-oxygen bottom waters on the LA shelf are largely unsuitable to sustain many types of marine organisms,
39 threatening the local ecosystem and therefore endan gering the lucrative fishing industry in the Gulf of Mexico (Rabalais & Turner 2 001a, Adams et al. 2004). Strong relationships exist between the onset and ex pansion of the hypoxic zone and a substantial increase in the use of nitro gen-based commercial fertilizers within the Mississippi River drainage b asin beginning in the 1950Â’s (Goolsby & Battaglin 2001, Goolsby et al. 2001, Tur ner et al. 2007). It has also been well documented that the nitrate flux from the Mississippi River to the Gulf of Mexico has increased over the twentieth century (Goolsby et al. 1999, Alexander et al. 2000). This flux is a good indicat ion of the quantity of nitrogen being discharged from the Mississippi River. The de livery of such large amounts of freshwater, sediment and nutrients through the M ississippi River plume dramatically affects the biological and biogeochemi cal processes on the continental margin. Results from multiple studies h ave shown that nutrients delivered by the Mississippi River, along with stra tification, are a dominant factor controlling hypoxia on the Louisiana shelf (Rabalai s et al. 1996, Lohrenz et al. 1997, Rabalais et al. 1999, Scavia et al. 2003, Tur ner et al. 2006). Geochemical analyses of the sedimentary record can be used to investigate hypoxic conditions prior to monitoring data and effectively evaluate the associated environmental changes. This study ut ilizes multiple geochemical proxies on three sediment cores from the outer LA s helf. Several cores are necessary to provide a regional perspective of hypo xic conditions and the possible environmental stressors. By coring the out er shelf, we can examine not only the spatial extent of hypoxia but also observe the amount of impact of
40 nutrients on the shelf environment. Although the ge ochemical and biological processes in environment are inherently complicated we provide some possible explanations that could explain the observed trends Figure 10: Map showing core locations for cores used in this s tudy Methods Three sediment cores were taken from the outer LA s helf 150 to 200 km from the mouth of the Mississippi River (Figure 9, Table 4). Chronology was established by researchers at the USGS for each of the three cores using excess 210Pb (Osterman et al. 2005, Osterman et al. 2008, Swa rzenski et al. 2008). This suite of cores was selected to investigate the curr ent outermost extent of the hypoxic zone, and determine how far out on to the s helf hypoxic events extended in the past. Two cores (PE0305-GC1 and PE0305-BC1) capture last century and
41 are located within the chronic hypoxic zone as outl ined by (Rabalais et al. 1999). These cores are located in deeper water on the shel f where mid-summer hypoxia occurs 25 to 50% of the time. Core ID PE0305-GC1 PE0305-BC1 MRD05-BC6 Core Type Gravity Box Box Date Collected 7/28/02 7/28/02 5/12/05 Latitude 28 23.796 28 23.975 28.3970 Longitude 90 27.701 90 27.170 90.7091 Depth (mwd) 47 47 65 Length (cm) 164 31 42 Frequency of the Occurrence of Hypoxia 25 to 50% 25 to 50% Less than 25% 210Pb rate (cm/yr) 0.33* 0.34* 0.36+ published in (Osterman et al. 2005)*,(Osterman et a l. 2008)+, and #Rabalais et al. (2002b) Table 4: Information on the three sediment cores used in th is study Another sediment core (MRJ05-BC6) was taken from de eper water depth and also represents the most recent century. This s ite was selected because it is
42 located outside of the chronically hypoxic zone and can be used as a baseline environmental comparison to the other two hypoxic c ores (Figure 9). The analyses of this core should reflect an environment that is not affected by lowoxygen bottom water conditions. A low-oxygen tolerant faunal assemblage, known as t he PEB index, is a faunal proxy for low-oxygen conditions (Osterman et al. 2000, Osterman 2003). The relative abundance of a few opportunistic foram iniferal species ( Pseudononion atlanticum, Epistominella vitrea and Buliminella morgani ) that can tolerate short intervals of low-oxygen conditio ns will become relatively higher during hypoxic events than during normoxic conditio ns. PEB index proxy data from previously published studies on these cores in dicate that all three cores are affected by low-oxygen conditions in the past few d ecades (Figure 10) (Osterman 2003, Osterman et al. 2005, Osterman et al. 2008, S warzenski et al. 2008). Even the core in deepest water (MRD05-BC6) which was ori ginally thought to be completely unaffected by hypoxic conditions appears to have some indication of low-oxygen events, although not at the same magnitu de as the other two cores. The PEB index in the two shallower cores, PE0305-GC 1 and PE0305-BC1, increase at approximately the same rate, while MRD0 5-BC6 begins to increase earlier (1930) than the other two and at a lower ra te. PE0305-GC1 has the highest values in the most recent decades but does not begin to increase until about 1960. The other hypoxic core, PE0305-BC1 has values that are almost as high, but begins to increase just before 1940.
43 Figure 11: PEB index data for the three cores in this study (O sterman et al. 2008). Note that the core from deeper water, MRD05BC6, does show an increase in the PEB index, suggesting that this core also experiences hypoxic conditions, but to a lesser ext ent than the other two cores. A combination of bulk organic geochemical sedimenta ry analyses including total organic carbon (TOC), total nitroge n (TN), stable isotopic ratios of carbon ( 13C) and nitrogen ( 15N), and C:N ratios were completed to provide an assessment of the changing sources of organic matte r and insight into the environmental response to the onset and increasing severity of both recent and past low-oxygen events. Stable isotopes of nitrogen ( 15N) are often used to
44 evaluate the sources and source changes of nitrate in nutrients and particulate matter in aquatic systems (Chang et al. 2002, Panno et al. 2006), as well as the effects of biological processes on organic matter ( Lohrenz et al. 1990, Mehnert et al. 2007). Concentrations of total organic carbon in each samp le were determined using a UIC Carbon Coulometer. Bulk sediment isotop e analyses ( 13C and 15N) were run in duplicate using a Carlo Erba 2500 Se ries I elemental analyzer coupled with a continuous-flow Finnigan Mat Delta P lus XL stable-isotope mass spectrometer. Each sample was first homogenized and then acidified with 0.1N HCl in a glass beaker to remove all inorganic carbo nates. The samples were analyzed relative to a reference gas (CO2 for carbon and N2 for nitrogen) to report isotopic values in standard per mil notation relative to PDB for carbon isotopes and air international standards for nitrog en. Instrumental accuracy was monitored through analysis of known amounts of spin ach leaf standards at regular intervals. Average standard deviations of i sotopic measurements for the standards were 0.17Â‰ for nitrogen and 0.21Â‰ for car bon in this data set. Results Geochemical results indicate that environmental cha nges on the LA shelf are recorded in the sediment record. TOC concentrat ions are increasing in all cores and range between 0.495% and 1.96% (Figure 11 ). This matches percentages of organic carbon in sediment cores fro m other areas of the shelf (Eadie et al. 1994, Swarzenski et al. 2008) PE0305GC1 and MRD05-BC6
45 contain similar amounts of TOC, and both show small fluctuations (approximately 0.3% and less). The amount of organic carbon in bot h cores is fairly stable until the TOC records in both begin to increase at around 1910. Core PE0305-BC1 contains roughly double the amount of organic carbo n as the other two cores, and increases from its oldest date around 1910 unti l 1980 when it decreases. This core also shows lager-scale variations (approx imately 0.5% or greater) than the other two cores. The tops of all cores have 13C values that show that the carbon is mostly from in situ phytoplankton sources (-22Â‰) rather than from terre strial plant sources (-27Â‰) brought in from the continent by the Mississippi River. Both PE0305-GC1 and PE0305-BC1 have somewhat more deplet ed carbon isotopes before 1950, suggesting that terrestrial input was more of a factor in these two cores in the past. The bottom of PE0305-BC1 has val ues around -24Â‰ until about 1940, where values change from less than -23Â‰ to greater than -22Â‰. PE0305-GC1 contains several fluctuations in which v alues are greater than 26Â‰, and begins to decrease at around 1930. In cont rast, the deeper water depth core, MRD05-BC6, remains constant at -22Â‰ thr oughout the length of the record, suggesting that the source of the organic c arbon in this core has not changed. C:N ratios of the same material also reflect algal sources, and these values seem to show less change than 13C in each of the three cores. PE0305BC1 and MRD05-BC6 have similar values that remain u nchanged throughout the length of the records at around 9, indicative of al gal sources. PE0305-GC1
46 mostly has values centered around 11, with some exc ursions around 17. The largest of these excursions occur at 1900 and 1970. The percent of nitrogen is increasing in each of th e three cores from about 0.048% to about 0.15%, which is approximately a thr ee-fold increase (Figure 12). PE0305-GC1 contains the least amount of nitrogen, w hile PE0305BC1 contains the most. Overall, these values appear to be increa sing at approximately the same rate. In PE0305-GC1, the %N remains relatively constant until about 1900, at which point the amount of nitrogen increases to almost double. In MRD05BC6, percent nitrogen increases steadily until 1930 when the rate of increase is slightly higher than before. Nitrogen percentage in creases in PE0305-BC1 appears to be constant, with smaller, decadal scale variations. In all sediment cores, 15N values progressively increase towards the top of the record, although at different rates. The val ues at the bottom of PE0305GC1 and MRD05-BC6 start at around 3.5-4Â‰ and increa se steadily to 5.5-6Â‰ at the top of the cores. PE0305-BC1 has a value of 4Â‰ at the bottom and a value of around 6Â‰ at the top. This core also has the sharpe st increase of about 3Â‰ that occurs just after 1940. In addition, the bulk 15N values found in the most recent portions of the sediment cores agree with the range found in Mississippi River particulate organic matter of 5-9Â‰ (Battaglin et al 2001, Kendall et al. 2001) and other sediment cores (Eadie et al. 1994). Although the overall change in nitrogen isotopic values is roughly the same in each core (a pproximately 2.5Â‰) from
47 Figure 12: Geochemistry (TOC, 13C, C:N) results from each of the three cores. The core from deepest water, MRD05-BC6, is d epicted in the darkest color, and the two cores affected by hy poxic conditions, PE0305-GC1 and PE0305-BC1 are shown in lighter colo rs. Raw data are included in Appendix A.
48 bottom to top, the rates of change differ between c enturies. During the nineteenth century, values increased about 0.5Â‰ (3Â‰ to 3.5Â‰ in PE0305-GC1 and 3.75Â‰ to 4.25Â‰ in MRD05-BC6). PE0305-BC1 is unique in tha t the values at the bottom of that core decrease to about 2.5% until th e mid-1940Â’s when the values begin to rapidly increase to a value of 6Â‰. Discussion Analysis of three sediment cores from the Louisiana continental shelf indicates that anthropogenic nutrient loading has s ignificantly impacted the local marine environment within the hypoxic zone. Increas ing TOC values in all cores reflect hypoxic conditions through either increased preservation (Demaison & Moore 1980) or increased primary productivity in th e surface waters exporting increased amounts of organic matter to the sediment s (Calvert & Pedersen 1992). Stable carbon isotopic ratios from all three cores suggest that a majority of the organic matter in sediments is in situ phytoplankton in origin. The two cores from an area of 25 to 50% summertime hypoxia indica te that there has been a change in the predominant source of organic carbon from terrestrial to algal. These results could be due to increasing amounts of in situ phytoplanktonderived inputs with stable terrestrial inputs, or r elatively stable algal inputs and decreasing amounts of terrestrial organic matter in put. Stable isotopes of carbon in other cores from the LA shelf also show a transi tion from values representative of terrestrial material to more algal values (Eadie et al. 1994). Much of the
49 organic carbon preserved during periods of hypoxia likely originates from increased production of algal material in response to nutrient loading, in addition to a smaller percentage of terrestrially-derived or ganic material. MRD05-BC6 has 13C and C:N values that are indicative of a primarily algal source throughout the length of the core. Although this core has a PEB in dex indicative of low oxygen since about 1930, the PEB indices dating back to 18 40 to not indicate low oxygen (Osterman et al. 2008). MRD05-BC6 is located in deeper water than the two other cores and farther from the influence of t he Mississippi River terrestrially-derived organic carbon. If this core has not historically been as influenced by Mississippi River discharge or hypoxi a, then other factors must be driving the change in the nitrogen isotopes, not ju st in the deeper water core but in the other two as well. The trend in 15N could not be the result of diagenetic processes. In fact, a strictly diagenesis signal would produce a trend ex actly opposite of what is observed. Sedimentary bacterial processes preferent ially utilize and remove the lighter isotope of nitrogen ( 14N) first, leaving behind the heavier isotope ( 15N) (Berner 1971, Altabet 2006). This removal of the li ghter isotope causes the 15N record to become more enriched downcore as more and more of the ligh t isotope is removed and utilized. Because this is exactly op posite of the isotopic trend observed in the cores in this study, this record do es not simply reflect a diagenetic signal, but is indicative of changing ni trogen inputs to the sediments on the shelf.
50 Figure 13 : Stable nitrogen isotopes and percent nitrogen re sults for the three cores. The deeper water core is shown in the darkes t color. Raw data are included in Appendix A.
51 If diagenesis is not causing the enriched 15N in the cores, then there must be other geochemical and biological processes occurring that affect nitrogen isotopic ratios in shelf sediments. Biogeo chemical processes in the water column and in sediments can substantially inc rease the 15N value with increasing distance from the source. This trend in nitrogen stable isotope values could be caused by a multitude of factors. Fortunat ely, there are a limited number of sources that would cause heavy nitrogen isotopic values in sediments. In addition, the combination of several nitrogen sourc es might be a more probable explanation of the nitrogen isotopic values in thes e cores. Excess primary production is surface waters on the LA shelf is likely responsible for heavier sediment nitrogen values. P rimary production has been shown to be increasing in direct correlation with i ncreasing nutrient inputs into the northern Gulf of Mexico (Lohrenz et al. 1990, Turne r & Rabalais 1991, Lohrenz et al. 1997, Dagg & Breed 2003). The increasing amount s of TOC correspond with increasing percentages of nitrogen in the three sed iment cores, which can be considered a record of productivity (Altabet 2006). This supports the idea that primary production is increasing in this region. Ad ditionally, results from bulk geochemistry ( 13C, C:N, 15N) strongly suggest an increase of algal organic matter in the sediments, further supporting the the ory that primary production likely increased in response to nutrient loading. W here there is a sufficient supply of nutrients, biological uptake discriminates betwe en the heavy and light isotopes of nitrogen, tending to favor the incorporation of 14N over 15N. (Battaglin et al.
52 2001, O'Reilly & Hecky 2002). Lake studies have sho wn that fractionation during uptake by phytoplankton can be -4 to -5Â‰, meaning t hat the nitrogen source would have to be at least 4Â‰ heavier than the produ ced organic matter (Fogel & Cifuentes 1993). At first, the organic matter produ ced by primary producers is depleted relative to the nutrient source pool, and the remaining source is relatively enriched in 15N. As productivity continues, and the demand for nitrogen remains high, the demand for nitrogen woul d eventually outweigh the discrimination against the heavier isotope, and the heavier isotope is incorporated as nitrogen concentrations decline. So if productivity continues due to high nutrient availability and fractionation in this manner goes to completion, the producers eventually attain the same isotopic c omposition as their source (O'Reilly & Hecky 2002, Savage 2005). The enriched organic matter produced is eventually incorporated into sediments, and could r esult in the sedimentary 15N trends observed on the Louisiana shelf. These trends in nitrogen have been observed in othe r regions of the Louisiana shelf. Eadie et al. (1994) suggested that sediments on the shelf contain a local record of primary productivity. The y found that 13C and 15N values in two sediment cores near the outflow of th e MR show large changes beginning in the 1960Â’s that correspond with increa ses in nutrient loads from the MRB. Additionally, 15N values in those cores exhibit similar values and temporal trends as observed in the cores in this study, even though those cores are located on the shelf much nearer to the mouth of th e Mississippi River. Eadie et al. interprets these changes in 15N as evidence for increased primary
53 productivity on the shelf. Thus, increased primary production is a likely source of some of the shift observed in nitrogen isotopes. An alternative explanation for the increasingly hea vier sediment isotope values is that there could be a consistent source o f heavy 15N to the sediments. Mixing models can be a useful approach to determini ng source contributions as long as those sources have distinct isotopic signat ures (Savage 2005). Previous mass-balance approaches have found that the increas ed nitrogen load to the Louisiana shelf is due primarily to fertilizers and to a lesser extent, sewage and manure inputs (Goolsby et al. 1999, Alexander et al 2000, Savage 2005, Booth & Campbell 2007). However, commercially produced fe rtilizers are manufactured using the Haber-Bosch process which fixes atmospher ic nitrogen which, by definition, has a 15N ratio of 0 (Kendall 1998). Since agricultural fer tilizers are estimated to be the largest source of nitrogen to t hese sediments, the isotopic values should become increasingly lighter, not heavier as observed in these sediment cores. So, there must be other, heavier so urces of nitrogen contributing to the sediments in the Gulf of Mexico. Sewage and manure have much heavier isotopic value s than other nitrogen nutrient sources. Savage (2005) found sedi mentary 15N have approximately 3Â‰ over a 30 year period as a direct consequence of increasing 15N-enriched tertiary-treated sewage input into the B altic Sea. To cause similarly elevated 15N values in the northern Gulf of Mexico, a large in put of manure and sewage would be required if mixing of en riched sources were the only process affected the isotopic values (Panno et al. 2006). Such an input is
54 unrealistic, and clearly documented to be otherwise (Goolsby et al. 1999, Alexander et al. 2008), especially given that N loa d models estimate that only a relatively small percentage of NO3 comes from sewage or manure (Goolsby et al. 1999, Alexander et al. 2000, Booth & Campbell 2 007). Therefore, source change alone cannot be the only, or even a substant ial, factor influencing nitrogen isotopes in these sediment cores. If mixing of different sources of nitrate alone doe s not account for the changes in nitrogen isotopic values, there must be other processes, such as denitrification, that play an important role. This process can significantly affect stable isotope ratios of nitrogen. Denitrification is the microbially mediated process by which fixed nitrogen, specifically NO3, is reduced to N2 under low oxygen conditions (Battaglin et al. 2001, Mehnert e t al. 2007) and may be a significant marine sink of nitrogen (Devol 1991). A dditionally, because the denitrification process can have fractionations in the range of up to 10 to 30Â‰ (Wada & Hattori 1991), stable isotopes of nitrogen can effectively be used as a tool to estimate of the extent of denitrification ( Mehnert et al. 2007). Significant positive shifts in nitrogen isotope ratios can be a strong indication that denitrification is consuming NO3 (Kellman & Hillaire-Marcel 1998). A large percentage (44%) of the total (global) deni trification occurs on continental shelves and inshore waters, which are t he areas most impacted by anthropogenic activities (Seitzinger et al. 2006). For denitrification to occur on continental shelves, strong vertical stratification and large amounts of organic matter are necessary. Both of these conditions are met on the Louisiana shelf
55 during the seasonal periods of hypoxia, and hypoxic conditions likely enhance the denitrification process (Dagg & Breed 2003). Th us, it is likely that nitrogen isotopes are influenced by denitrification on the L ouisiana continental shelf. On seasonally hypoxic continental shelves, such as the LA shelf, nutrient inputs are distributed throughout the water column by vertical mixing, making them available for denitrification when the water c olumn becomes stratified and dissolved oxygen concentrations drop (O'Reilly & He cky 2002, Seitzinger et al. 2006). As nitrogen loading increases, there is more potential for denitrification (Seitzinger et al. 2006). Since nitrate loading on the LA shelf has been increasing over several decades (Goolsby et al. 1999, Goolsby et al. 2001) it stands to reason that the rate of denitrification on the shel f has also most likely increased. Additionally, denitrification rates were found to i ncrease with increasing amounts of organic carbon in the sediments on the LA shelf (DeLaune et al. 2005). Although the amount of organic matter in the cores in this study is only approximately 1%, there is enough organic matter pr esent to encourage denitrification in the sediment. Another possibility is that a substantial amount of denitrification is occurring within the Mississippi River Basin. Denit rification that takes place within the Mississippi watershed or within the Mississippi River itself would enrich the source pool of nitrogen that eventually is discharg ed to the Gulf of Mexico. Several studies have suggested that much of the den itrification that takes place in the Mississippi watershed occurs before discharg e into the River (Kendall
56 1998, Panno et al. 2006, Seitzinger et al. 2006) an d is probably minimal within the Mississippi River itself (Battaglin et al. 2001 Panno et al. 2006). Denitrification in the Mississippi basin is most li kely fueled by the runoff of nutrients applied to agricultural fields. Approxima tely 58% of the land in the Mississippi River drainage basin is used for agricu lture. Much of this land is currently drained by tile drainage systems, which e mploy perforated tiles or pipes to contour the fields and move excess water into dr ainage ditches. This system allows farmers to farm lands that are usually too w et for agricultural use. Although these tiles encourage nutrient runoff from fields and directly supply large quantities of nutrients to the river (McIsaac & Hu 2004), they are not thought to be a major source of denitrification (Me hnert et al. 2007). Tile drains reroute water carrying nutrients that would normall y pass into subsurface soil layers, where a majority of denitrification is thou ght to take place (Mehnert et al. 2007). Water samples from the Mississippi River and tile d rains within the Mississippi basin have been found to have 15N values between 4.8 and 16.4Â‰ (Panno et al. 2006). Values from tile drains were f ound to be at the lower end of the scale, with an average in one study being 5.2Â‰, while values in groundwater wells were generally >12Â‰ (Mehnert et al. 2007), su ggesting that more denitrification is occurring in subsurface groundwa ter rather than tile drains. The average of tile drainage values and groundwater val ues that flow into the Mississippi River is probably in the order of 5 to 9Â‰, which is the isotopic values of particulate organic matter in the River found by other studies (Battaglin et al.
57 2001, Kendall et al. 2001). If this is the isotopic value of the particulate organic matter that is discharged to the Gulf of Mexico, an d this particulate matter is incorporated into shelf sediments, it could be that denitrification in the Mississippi basin alone could be responsible for the isotopic v alues observed in LA shelf sediments. Summary Increasingly enriched nitrogen isotope sediment va lues on the LA shelf were observed in three sediment cores. Geochemistry demonstrates that while two of these cores have undergone a transition from receiving organic matter from terrestrial sources to having primarily algal sources, one of these cores has always been influenced primarily by algal sources. This core has remained unchanged in every proxy except that it demonstrate s increasing nitrogen stable isotopes, which may be the key to understanding how the environmental is changing on the LA shelf. Since diagenetic processe s are not likely responsible for this trend, it is probably the result of severa l inter-related biological and geochemical processes. Changing source inputs, part icularly sewage and manure which contain enriched nitrogen isotopes, ar e likely increasing due to increasing population and increased agricultural ac tivity in the Mississippi River basin. However, these inputs are not significant en ough to be the only, or a substantial, factor driving 15N values to be more enriched. The denitrification process, which results in enriched isotopic values, is probably a significant factor both within the Mississippi River basin and on the LA shelf.
58 Excess primary production on the LA shelf could a lso be responsible for more enriched sediment nitrogen values. Where there is a sufficient supply of nutrients, biological uptake discriminates between the heavy and light isotopes of nitrogen. As primary production continues, and the demand for nitrogen remains high, the demand for nitrogen would eventually outw eigh the discrimination against the heavier isotope, and the heavier isotop e is incorporated into biological material as nitrogen concentrations decl ine. This enriched organic matter would eventually be incorporated into shelf sediments. Increasing nitrogen flux to the Gulf of Mexico (Goolsby et al. 1999, Go olsby et al. 2001) corresponds temporally with increases in sedimentary total nitr ogen on the LA shelf. Since it has been reasonably proven that primary production has increased with corresponding increases of river-borne nutrients (E adie et al. 1994, Lohrenz et al. 1997, Rabalais et al. 2002a), it is very possible t hat an increase in primary production responsible for the increasing 15N values in the sediment cores is related to increased primary production and not oth er processes. For the isotopic records for nitrogen to steadily increase over time, there must be an increase in the rates of denitrification or production, and change in the source of the nitrogen brought on to the LA she lf, or a combination of several of these factors. For a more detailed view of the e cosystem, geochemical analyses of specific biomarkers are needed.
59 Chapter 4: A Multi-Proxy Sedimentary Record of Hist orical LowOxygen Conditions on the Louisiana Continental Shel f Introduction The Louisiana continental shelf is chronically affe cted by seasonal hypoxia during spring and summer, and is currently one of t he worldÂ’s largest zones of coastal human-caused hypoxia (Rabalais et al. 2002c ). Every year, large quantities of nutrients from the continent are deli vered via the Mississippi River to the Gulf of Mexico, causing large annual production blooms. Based on field observations conducted since 1985, both the severit y of oxygen depletion and the spatial extent of the chronically hypoxic zone on the LA continental shelf have increased significantly (Rabalais et al. 1994, Raba lais et al. 2001, Rabalais et al. 2007b). It is generally accepted that hypoxic conditions to day on the LA shelf have been exacerbated by anthropogenic nutrient inputs. Annual monitoring of the hypoxic zone did not begin until 1985, and as such, a record that describes the marine ecosystem environment in detail prior to rec ent monitoring does not exist. The sediment record contains paleoindicators of lon g-term ecological and environmental changes related to low-oxygen conditi ons on the LA continental shelf. Sediment cores from inside the current hypox ic region contain chemical and biological remains that reflect environmental c onditions in surface and
60 bottom waters at the time the sediments were deposi ted and provides clues of historic changes (Rabalais et al. 2002c). The onset and expansion of the hypoxic zone is larg ely attributed to anthropogenic influences, particularly an increase in the use of nitrogen-based commercial fertilizers within the Mississippi River drainage basin beginning in the 1950Â’s (Nelsen et al. 1994, Goolsby & Battaglin 200 1, Goolsby et al. 2001). However, hypoxic events have been shown to have occ urred in the past prior to anthropogenic nutrient input, and that some of thes e events have been similar in magnitude to the most recent low-oxygen conditions (Turner & Rabalais 1994, Osterman 2003, Osterman et al. 2005). These studies have suggested that Mississippi River discharge is the driving factor c ausing historic low-oxygen conditions on the LA shelf prior to 1950. These res ults indicate that hypoxia may be a natural, aperiodic phenomenon on the LA shelf driven by the Mississippi River, and that this process has been exacerbated b y excess nutrient input to cause the Â“modernÂ” hypoxic event. The historic record provides a view of a distinct c ontrast between recent hypoxic conditions and environmental conditions in the past. Previous sediment core studies on the LA shelf clearly document (thro ugh marine-origin carbon evident by stable carbon isotope analysis) recent e utrophication and increased organic sedimentation, with the change more apparen t in areas of chronic hypoxia and coincident with the increasing nitrogen loads from the Mississippi River system beginning in the 1950Â’s (Eadie et al. 1994). Determining the natural and anthropogenic factors that control the formatio n of low-oxygen conditions on
61 the LA shelf is crucial to understanding how the mo dern Â“Dead ZoneÂ” develops. To do this, a historical record of past hypoxic eve nts on the shelf needs to be determined. Several studies have attempted to look at ecosystem and environmental changes in the northern Gulf of Mexic o in conjunction with hypoxia (Sen Gupta et al. 1996, Chen et al. 2001, Rabalais et al. 2004, Chen et al. 2005, Osterman et al. 2008, Swarzenski et al. 2008). The goal of this research is to use a multi-proxy geochemical approach to extend the re cord of LA shelf hypoxia into the past prior to 1985 monitoring. Methods A multi-proxy geochemical approach was used to reco nstruct environmental conditions associated with historical low-oxygen events on the LA shelf. A gravity core was collected within a region of 25 to 50% occurrence of mid-summer hypoxia. This sediment core, PE0305-GC1, was collected in 47 meters of water during a USGS cruise in 2003 (Figur e 13). The core (1.64m in length) was sampled every centimeter. Chronology wa s established by researchers at the USGS for the upper portion of th e core using 210Pb (Osterman et al. 2005).
62 Figure 14: Map showing core location relative to frequency of hypoxic conditions on the LA shelf for the period 1985-2002 (from Rabalais et al. 2002a) A low-oxygen tolerant faunal assemblage, known as t he PEB index, is a foraminiferal proxy for relative conditions of oxyg en in bottom waters. As benthic oxygen concentrations drop, many species of foramin ifers will die out. Therefore, the abundance of a few opportunistic species that c an tolerate short intervals of low-oxygen conditions will become relatively higher than during normoxic conditions. The relative percentage increase of thr ee benthic foraminiferal species ( Pseudononion atlanticum, Epistominella vitrea and Buliminella morgani ) in Louisiana shelf sediments is used as a proxy f or past low-oxygen conditions, although it cannot determine the precis e value of oxygen concentration. Previously published studies of the PEB index in this core have
63 identified that it documents several pre-anthropoge nic low-oxygen events (Osterman 2003, Osterman et al. 2005, Osterman et a l. 2008). Chronology in the lower portion of PE0305-GC1 is pr oblematic. 210Pb can be accurately measured in sediments and used to est imate age to within approximately the past 100 years (Bierman et al. 19 98). However, 210Pb cannot be used to date sediments that are older than 100 y ears. A date of approximately 1900 is shown on Figure 15. Portions of the core be low this point were not able to be dated, so the timing, duration and amount of time between low-oxygen events are uncertain. Additionally, the LA continen tal shelf is subjected to several factors that affect and disrupt sedimentation (Alli son et al. 2000, Corbett et al. 2006). As a result, it can be assumed that at least some of the lower portion of this sediment core has not remained intact. Also, i t is very probable that lowoxygen events have occurred elsewhere on the LA she lf that are not evident in this core. A combination of bulk organic geochemical sedimenta ry analyses, such as total organic carbon (TOC), stable isotopic ratios of carbon ( 13C) and nitrogen ( 15N), and C:N ratios, were completed to provide an as sessment of the changing sources of organic matter and insight into the environmental parameters that might be related to changes in oxyg en conditions over time. Concentrations of total organic carbon in each samp le were determined using a UIC Carbon Coulometer at the University of South Fl orida. Bulk sediment isotope analyses ( 13C and 15N) were run in duplicate using a Carlo Erba 2500 Se ries I elemental analyzer coupled with a continuous-flow F innigan Mat Delta Plus XL
64 stable-isotope mass spectrometer. Each sample was f irst homogenized and then acidified with 0.1N HCl in a glass beaker to remove all inorganic carbonates. The samples were analyzed relative to a reference gas ( CO2 for carbon and N2 for nitrogen) to report isotopic values in standard per mil notation relative to PDB for carbon isotopes and air international standards for nitrogen. Instrumental accuracy was monitored through analysis of known am ounts of spinach standards at regular intervals. Average standard de viations of isotopic measurements for the standards were 0.17Â‰ for nitro gen and 0.21Â‰ for carbon in this data set. Results TOC In the presence of hypoxic conditions, total organ ic carbon in sediments increases due to increased preservation (Demaison & Moore 1980), increased primary productivity (Calvert & Pedersen 1992) or a combination of both. In PE0305-GC1, TOC values vary in different sections o f the core, but ranges between 0.37 Â– 1.03%, with a mean value of 0.68% (F igure 14). Overall, the highest values of TOC were found in the upper 30 cm of the core, and the lowest values were found in the bottommost part of the cor e. Values in the lowest portion of the core increase from around 0.4% to ab out 1.0%, were found in the lowest 130 to 164 cm of the core. The intermediate sections are a relatively stable 0.7%. The average value decreases slightly t o about 0.6% at 90 cm core depth until about 30 cm. Between 30 cm and the top of the core the amount of
66 Figure 15 : Geochemistry results for PE0305-GC1. PEB index re sults are also included and have been published previously by Oste rman et al. (2005). A date of 1900 makes the approximate limit in the core of 210Pb dating. Letters denote PEB events that are great er than the mean value for the overall core. Gray bars indicate low-oxygen events in the PEB that correspond with geochemical indicators of terrestrial organic matter. The raw data used in th is figure are listed in Appendix A. organic carbon steadily increases to around 1.0%. A dditional episodic high and low values were observed in various parts of the co re. Carbon Isotopes Stable isotopes ratios of carbon ( 13C) at the bulk sediment level provide a first-order evaluation of the source of sedimentary organic matter (Meyers 1994). Different carbon sources have distinct isotopic val ues, allowing terrestrial material (-27Â‰) to be differentiated from marine-de rived material (-21Â‰) in marine systems (Killops & Killops 2005). 13C values in this core range between -20Â‰ and -28Â‰. In addition to small-scale fluctuati ons, there are several largescale changes between more depleted (terrestrial) v alues and more enriched (algal) values. In the uppermost and lowermost port ion of the core, 13C values show that the carbon is increasingly from algal sou rces rather than from terrestrial plants from the continent. The middle p art of the core has mostly more depleted values, suggesting that the organic carbon in the sediments is mostly from terrestrial sources. Some enriched small-scale fluctuations in 13C correspond with TOC episodic peaks, suggesting a li nk between terrestrial source and increased organic matter, at least in so me portions of the core.
67 Nitrogen Isotopes Stable isotopes of nitrogen ( 15N) can be used to discriminate between trophic levels in ecosystems (O'Reilly & Hecky 2002 ), to evaluate the source and cycling of nitrate in sediments (Macko et al. 1993, Kendall 1998, Chang et al. 2002, Panno et al. 2006), or changes in environment al conditions over time (Meyers 1997). 15N values in PE0305-GC1 range between 1.94Â‰ and 5.86 Â‰ and average 3.63Â‰. Highest 15N values are observed at the top of the core, while the rest of the record tends to stay between 3Â‰ and 4Â‰. There are also few episodic peaks, as are observed in the 13C record, which also correspond to peaks observed in 13C. These peaks are not as distinct or as great in a mplitude in this record as they are in the carbon isotopic r ecord. At approximately 50 cm depth, 15N values begin to steadily increase to almost 6Â‰ near the core top. C:N Ratios Molar ratios of carbon to nitrogen (C:N) are also u seful indicators of organic matter source, as algal organic matter (4-1 0) and terrestrially organic matter (>20) have distinct signatures which can be detected in the sediment record (Jasper & Gagosian 1990, Meyers 1994). C:N r atios in PE0305-GC1 range from 8.3 to 20.7 and averages 12.3. Although C:N ratios are fairly constant throughout the core, there is a slight shift to low er values above 100 cm core depth. Above this depth, values are fairly consiste nt at 10, with some episodic peaks at 60, 40, and 15 cm depth. Below 100 cm, C:N ratios are slightly higher,
68 between 12 and 16. Below 150 cm, values are near 10 and below, which is similar to values observed at the top of the core. Some episodic peaks are also observed below 80 cm. Discussion While the most recent human-caused hypoxia is well characterized by instrumental observations and paleoindicators, much less is known about historic low-oxygen events. Geochemical analyses of a sedime ntary record from the LA shelf document recent hypoxic conditions in the nor thern Gulf of Mexico, as well as environmental changes that have occurred prior t o anthropogenic influences. The shifts observed in the 13C record suggest a distinct change in the source of organic matter at 30 cm. The section of the core be tween 30 and 110 cm shows increasingly lighter isotopic ratios, which reflect s an increasingly larger proportion of terrestrial material than in other sections of t he core. C:N ratios are also indicative of algal precursors, with some excursion s of higher values typically associated with terrestrial plant material. These c hanges could be related to climatic variability, as more terrestrial material must have been delivered by the Mississippi River from the North American continent during periods of peak discharge when shifts in landscape change were acce lerating in the middle of the country. Several PEB events prior to 1900 have been identif ied by Osterman et al. (2003) and Osterman et al. (2005), as designated by letters in Figure 15, based the values above the mean for the entire core. Some of these events are located
69 at the same core depths as the episodic peaks obser ved in the geochemical analyses, suggesting a relationship between the PEB low-oxygen index and the geochemical data. Several distinct episodic peaks a re evident in the TOC, 13C and C:N records, and are indicated with gray bars i n Figure 15. These peaks are consistent with input of more terrestrial material. Although these peaks vary in size between peaks and among proxies, they temporal ly correspond to each other. Geochemical indicators support the conclusio ns of other studies that lowoxygen events prior to the period of anthropogenic hypoxia were related to increased Mississippi River discharge because indic ators point to increased terrestrially-derived carbon at the same time. Three distinct windows of behavior are apparent in the geochemical data (Figure 15). During the upper 30 cm of the core, wh ich corresponds to the time period after 1950, there is a distinct relationship between the increasing PEB index, increasing TOC and shifting of 13C and C:N ratios to predominantly algal organic matter. This is the anthropogenic time wind ow, which covers the past 50 years and the most recent hypoxic event, and is cha racterized by rapidly worsening hypoxic conditions, exacerbated by anthro pogenic influences.
71 Figure 16 : Suggested time windows based on geochemical resul ts. The anthropogenic window occurs after 1950, the transit ion window between 1800 and 1950, and the historic window prio r to 1800. 13C, 15N and C:N records have been smoothed with a 5-point running average from the raw data (shown in gray) t o highlight trends. A Â“transition periodÂ” exists between 30 cm and 70 c m. Although reliable chronology is not available for the lower portion o f this core, if the sedimentation rate for the upper portion is extrapolated for the remainder of the core, 70 cm is approximately equivalent to 1800. This was the peri od of time after which the Louisiana Territory was purchased and intensive set tlement of the Mississippi River basin began. At this point in the core, TOC a nd 13C values begin to steadily increase. The timing of this change also c orresponds to increases in the amount of farmland in the basin, increased land cle aring (Turner & Rabalais 2003) and increases in sedimentary biogenic silica, a paleoindicator that may reflect in situ marine diatom density in response to organic N loa ds (Turner & Rabalais 1994). Nitrogen isotopic values begin to i ncrease at 50 cm depth, which would correspond to approximately 1850. This transi tion zone may reflect the natural ecosystem state beginning to be influenced by anthropogenic activities. During this period, 13C and C:N records begin to shift towards more marin e values and the PEB index events begin to decrease i n magnitude. Additionally, values in 15N values shift to more depleted values after 1800 a nd then begin to become more enriched. A third window, the historic window, exists below 7 0 cm and exhibits different behavior in all proxies. 13C values are similar to the anthropogenic
72 window, C:N ratios are more elevated, and TOC varie s during this window. A slight increase in 15N also occurs in the lowermost portion of the core, but values are not as high as observed during the anthr opogenic window or transition period. This time period is the only window of time that does not have any anthropogenic influence, and thus represents the na tural state of the ecosystem. It is also possible that some low-oxygen events occ urred elsewhere on the LA shelf but are not recorded in these cores. Recen t monitoring studies on the Louisiana shelf have found hypoxic conditions occur most commonly in water depths of 5-30 meters, but has been documented in d eeper depths as well (up to 60 m, but mostly less than 45 m) (Rabalais et al. 1 998, Rabalais et al. 1999). Osterman et al. (2008) used a transect of three cor es extending from the Mississippi River delta to show that more historic low-oxygen events occur in a core located in shallower water on the shelf and ne arer to the mouth of the Mississippi River where nutrient-enhanced primary p roduction is more likely to occur. PE0305-GC1 is located in 47 m water depth an d far enough out on the shelf that seasonal hypoxia does not occur in this region, only 25 to 50% of the time in mid-summer. Therefore, seasonally severe hy poxia that that occurs elsewhere on the LA shelf is not recorded in this s ediment core. The mechanism for these distinctive environmental categories is still unclear. It is possible that a combination of natur al climatic processes and anthropogenic activities such as land clearing and farming could be responsible for low-oxygen conditions prior to 1950 (Turner & R abalais 2003), and maybe even before then. Intensive settlement of the Midwe st began near the beginning
73 of the 19th century and included clearing of land primarily fo r agricultural purposes. These activities would have changed the a mount of sediment and organic matter brought into the Mississippi River a nd the Gulf of Mexico. A proposed mechanism for historic low-oxygen events has been suggested to be increased river flow (Osterman et a l. 2005). Changes in the PEB index before 1950 have been attributed to natural f luctuations, most likely peaks in discharge, in the Mississippi River (Osterman et al. 2005, Osterman et al. 2008, Swarzenski et al. 2008). My geochemical analy ses verify a prior terrestrial carbon signal at the time of these peaks that is su bsequently lost in the current anthropogenic epoch because of the increased anthro pogenic nutrient-enhanced in situ primary production.
74 Chapter 5: Summary Increasingly enriched nitrogen isotope sediment va lues on the LA shelf were observed in three sediment cores. Geochemistry indicates that while two of these cores have undergone a transition from receiv ing organic matter from terrestrial sources to having primarily marine sour ces, one of these cores has always been influenced by marine algal sources. Thi s core has remained unchanged in every proxy except that it demonstrate s increasing nitrogen stable isotopes, which may be the key to understanding how the environmental is changing on the LA shelf, but there are no studies to support this expected process. Since diagenetic processes are not likely responsible for this trend, it is probably the result of several inter-related biolog ical and geochemical processes. Since it has been reasonably proven that primary pr oduction and phytoplankton biomass has increased with corresponding increases of river-borne nutrients (Eadie et al. 1994, Lohrenz et al. 1997, Rabalais e t al. 2002a), it is very possible that an increase in primary production is at least partially responsible for the increasing 15N values in the sediment cores. Changing source inp uts, particularly sewage and manure contain enriched nit rogen isotopes, are likely increasing due to increasing population and increas ed agricultural activity in the Mississippi River basin. However, these inputs are minimal in the overall nitrogen and phosphorus inputs to the Gulf of Mexico and not significant enough to be a
75 factor driving 15N values to be more enriched. The denitrification p rocess, which results in enriched isotopic values, is probably a factor both within the Mississippi River basin and on the LA shelf, but there are no s tudies to support this expected process. The most likely explanation is that a comb ination of these factors is influencing nitrogen isotopes in sediments on the L A shelf. While hypoxia is well characterized by instrumental measurements since the early 1970s, paleoindicators have clearly docum ented changes in increased productivity in surface waters and deteriorating ox ygen conditions for benthos. Geochemical analyses of a sedimentary record from t he LA shelf help pinpoint the shifts in biological processes and sources of o rganic matter leading to a decline in dissolved oxygen concentrations. Three d istinct periods are apparent in the geochemical data. The anthropogenic time win dow, which covers the past 50 years and rapidly worsening hypoxic conditions, is associated with anthropogenic nutrient inputs. A transition period exists between approximately 1800 and 1950, which is the period of time in which the Louisiana Territory was purchased and intensive settlement of the Mississip pi River basin began. During this period, geochemistry indicates that a change h as occurred in the environment, and could be attributed to changing la nd use and increasingly intensive agriculture. The third window exists befo re 1800 and is the time period without obvious anthropogenic influence. The geochemical analyses of three sediment cores in dicates corresponding peaks with several PEB peaks prior to 1900 that indicate lower oxygen concentrations (Osterman 2003, Osterman et a l. 2005). The geochemical
76 characteristics of inputs of terrestrially-derived organic material match the PEB peaks before the anthropogenic hypoxia, indicating that pre-anthropogenic lowoxygen events were supported with higher amounts of organic matter from terrestrially-derived material deposited on the con tinental shelf during periods of peak discharge from the Mississippi River.
77 References Cited Adams CM, Hernandez E, Cato JC (2004) The economic significance of the Gulf of Mexico related to population, income, employment minerals, fisheries, and shipping. Ocean & Coastal Management 47:565-580 Alexander RB, Smith RA, Schwarz GE (2000) Effect of stream channel size on the delivery of nitrogen to the Gulf of Mexico. Nat ure 403:758-761 Alexander RB, Smith RA, Schwarz GE, Boyer EW, Nolan JV, Brakebill JW (2008) Differences in phosphorus and nitrogen deliv ery to the Gulf of Mexico from the Mississippi River Basin. Environmen tal Science & Technology 42:822-830 Allison MA, Kineke GC, Gordon ES, Goni MA (2000) De velopment and reworking of a seasonal flood deposit on the inner continenta l shelf off the Atchafalaya River. Continental Shelf Research 20:22 67-2294 Altabet MA (2006) Isotopic tracers of the marine ni trogen cycle: present and past. In: Volkman JK (ed) Marine Organic Matter, Vol 2. S pringer, Berlin, p 251293 Battaglin WA, Kendall C, Chang CCY, Silva SR, Campb ell DH (2001) Chemical and isotopic evidence of nitrogen transformation in the Mississippi River, 1997-98. Hydrological Processes 15:1285-1300 Berner RA (1971) Principles of Chemical Sedimentolo gy, Vol. McGraw-Hill Book Company, New York Bierman PR, Albrecht A, Bothner MH, Brown ET, Bulle n TD, Gray LB, Turpin L (1998) Erosion, weathering, and sedimentation. In: Kendall C, McDonnell JJ (eds) Isotope Tracers in Catchment Hydrology. El sevier, Amsterdam Biggs DC, Sanchez LL (1997) Nutrient enhanced prima ry productivity of the Texas-Louisiana continental shelf. Journal of Marin e Systems 11:237-247 Blackwelder P, Hood T, Alvarez-Zarikian C, Nelsen T A, McKee BA (1996) Benthic foraminifera from the NECOP study area impa cted by the Mississippi River plume and seasonal hypoxia. Quate rnary International 31:19-36
78 Booth MS, Campbell C (2007) Spring nitrate flux in the Mississippi River Basin: A landscape model with conservation applications. Env ironmental Science & Technology 41:5410-5418 Bratkovich A, Dinnel SP, Goolsby DA (1994) Variabil ity and prediction of freshwater and nitrate fluxes for the Louisiana-Tex as shelf: Mississippi and Atchafalaya River source functions. Estuaries 1 7:766-778 Burdige DJ (2006) Geochemistry of Marine Sediments, Vol. Princeton University Press, Princeton, NJ Calvert SE, Pedersen TF (1992) Organic carbon accum ulation in marine sediments: how important is anoxia? In: Whelan JK, Farrington JW (eds) Productivity, Accumulation, and Preservation of Org anic Matter in Recent and Ancient Sediments. Columbia University Press CENR (2000) Integrated assessment of hypoxia in the northern Gulf of Mexico, National Science and Technology Council Committee o n Environmental and Natural Resources, Washington, D.C. Chang CCY, Kendall C, Silva SR, Battaglin WA, Campb ell DH (2002) Nitrate stable isotopes: tools for determining nitrate sour ces among different land uses in the Mississippi River Basin. Canadian Journ al of Fisheries and Aquatic Sciences 59:1874-1885 Chen N, Bianchi TS, McKee BA (2005) Early diagenesi s of chloropigment biomarkers in the lower Mississippi River and Louis iana shelf: implications for carbon cycling in a river-dominated margin. Mar ine Chemistry 93:159177 Chen N, Bianchi TS, McKee BA, Bland JM (2001) Histo rical trends of hypoxia on the Louisiana shelf: application of pigments as bio markers. Organic Geochemistry 32:543-561 Cloern JE (2001) Our evolving conceptual model of t he coastal eutrophication problem. Marine Ecology Progress Series 210:223-253 Corbett DR, Dail M, McKee B (2007) High-frequency t ime-series of the dynamic sedimentation processes on the western shelf of the Mississippi River Delta. Continental Shelf Research 27:1600-1615 Corbett DR, McKee B, Duncan D (2004) An evaluation of mobile mud dynamics in the Mississippi River deltaic region. Marine Geo logy 209:91-112 Corbett DR, McKee BA, Allison M (2006) Nature of de cadal-scale sediment accumulation on the western shelf of the Mississipp i River delta. Continental Shelf Research 26:2125-2140
79 Dagg MJ, Breed GA (2003) Biological effects of Miss issippi River nitrogen on the northern gulf of Mexico a review and synthesis. J ournal of Marine Systems 43:133-152 David MB, Wall LG, Royer TV, Tank JL (2006) Denitri fication and the nitrogen budget of a reservoir in an agricultural landscape. Ecological Applications 16:2177-2190 DeLaune RD, Jugsujinda A, Devai I, Hou AX (2005) De nitrification in bottom sediment near oil production facilities off the Lou isiana Gulf Coast. Chemistry and Ecology 21:101-108 Demaison GJ, Moore GT (1980) Anoxic environments an d oil source bed genesis. The American Association of Petroleum Geol ogists Bulletin 64:1179-1209 Devol AH (1991) Direct measurement of nitrogen gas fluxes from continental shelf sediments. Nature 349:319-321 Diaz RJ (2001) Overview of hypoxia around the world Journal of Environmental Quality 30:275-281 Diaz RJ, Rosenberg R (1995) Marine benthic hypoxia: A review of its ecological effects and the behavioural responses of benthic ma crofauna. Oceanography and Marine Biology: an Annual Review 3 3:245-303 Donner SD (2003) The impact of cropland cover on ri ver nutrient levels in the Mississippi River Basin. Global Ecology and Biogeog raphy 12:341-355 Donner SD, Coe MT, Lenters JD, Twine TE, Foley JA ( 2002) Modeling the impact of hydrological changes on nitrate transport in the Mississippi River Basin from 1955 to 1994. Global Biogeochemical Cycl es 16:doi:10.1029/2001GB001396 Donner SD, Kucharik CJ (2003) Evaluating the impact s of land management and climate variability on crop production and nitrate export across the Upper Mississippi Basin. Global Biogeochemical Cycles 17:doi:10.1029/2001GB001808 Donner SD, Kucharik CJ, Foley JA (2004) Impact of c hanging land use practices on nitrate export by the Mississippi River. Global Biogeochemical Cycles 18:doi:10.1029/2003GB002093 Dortch Q, Rabalais NN, Turner RE, Rowe GT (1994) Re spiration rates and hypoxia on the Louisiana shelf. Estuaries 17:862-87 2
80 Eadie BJ, McKee BA, Lansing MB, Robbins JA, Metz S, Trefry JH (1994) Records of nutrient-enhanced coastal ocean producti vity in sediments from Louisiana continental shelf. Estuaries 17:754765 Ertle JR, Hedges JI (1985) Sources of sedimentary h umic substances: vascular plant debris. Geochimica et Cosmochimica Acta 49:20 97-2107 Fogel ML, Cifuentes LA (1993) Isotope Fractionation during Primary Production. In: Engle MH, Macko SA (eds) Organic Geochemistry. Plenum Press, New York, p 73-94 Fry B (2007) Stable Isotope Ecology, Vol. Springer, New York Galloway JN, Cowling EB (2002) Reactive nitrogen an d the world: 200 years of change. Ambio 31:64-71 Goni MA, Ruttenberg KC, Eglinton TI (1997) Sources and contribution of terrigenous organic carbon to surface sediments in the Gulf of Mexico. Nature 389:275-278 Goolsby DA, Battaglin WA (2001) Long-term changes i n concentrations and flux of nitrogen in the Mississippi River Basin, USA. Hy drological Processes 15:1209-1226 Goolsby DA, Battaglin WA, Aulenbach BT, Hooper RP ( 2001) Nitrogen input to the Gulf of Mexico. Journal of Environmental Qualit y 30:329-336 Goolsby DA, Battaglin WA, Lawrence GB, Artz RS, Aul enbach BT, Hooper RP, Keeney DR, Stensland GJ (1999) Flux and Sources of Nutrients in the Mississippi-Atchafalaya River Basin: Topic 3 Report for the Integrated Assessment on Hypoxia in the Gulf of Mexico NOAA Co astal Ocean Program, Silver Spring, MD Hedges JI, Parker PL (1976) Land-derived organic ma tter in surface sediments from the Gulf of Mexico. Geochimica et Cosmochimica Acta 40:1019-1029 Jasper JP, Gagosian RB (1990) The sources and depos ition of organic matter in the Late Quaternary Pigmy Basin, Gulf of Mexico. Ge ochimica et Cosmochimica Acta 54:1117-1132 Justi D (1991) Hypoxic conditions in the northern Adriat ic Sea: historical development and ecological significance. In: Tyson RV, Pearson TH (eds) Modern and Ancient Continental Shelf Anoxia, Vol 58 Geological Society Special Publication, London, p 95-105
81 Justi D, Legovi T, Rottini-Sandrini L (1987) Trends in oxygen cont ent 19111984 and occurrence of benthic mortality in the nor thern Adriatic Sea. Estuarine, Coastal and Shelf Science 25:435-445 Justi D, Rabalais NN, Turner RE (1997) Impacts of climat e change on net productivity of coastal waters: implications for ca rbon budgets and hypoxia. Climate Research 8:255-237 Justi D, Rabalais NN, Turner RE (2003) Simulated respons es of the Gulf of Mexico hypoxia to variations in climate and anthrop ogenic nutrient loading. Journal of Marine Systems 42:115-126 Justi D, Rabalais NN, Turner RE, Wiseman Jr. WJ (1993) S easonal coupling between riverborne nutrients, net productivity and hypoxia. Marine Pollution Bulletin 26:184-189 Kellman L, Hillaire-Marcel C (1998) Nitrate cycling in streams: using natural abundances of NO-3 d15N to measure in-situ denitrification. Biogeochemistry 43:273-292 Kendall C (1998) Tracing nitrogen sources and cycli ng in catchments. In: Kendall C, McDonnell JJ (eds) Isotope Tracers in Catchment Hydrology. Elsevier, Amsterdam, p 519-576 Kendall C, Silva SR, Kelly VJ (2001) Carbon and nit rogen isotopic compositions of particulate organic matter in four large river s ystems across the United States. Hydrological Processes 15:1301-1346 Killops S, Killops V (2005) Introduction to Organic Geochemistry, Vol. Blackwell Publishing Kohl DH, Shearer GB, Commoner B (1971) Fertilizer n itrogen: Contribution to nitrate in surface water in a corn belt watershed. Science 174:1331-1334 Lohrenz, Steven E., Fahnenstiel GL, Redalje DG, Lan g GA, Chen X, Dagg MJ (1997) Variations in primary production of northern Gulf of Mexico continental shelf waters linked to nutrient inputs from the Mississippi River. Marine Ecology Progress Series 155:45-54 Lohrenz SE, Dagg MJ, Whitledge TE (1990) Enhanced p rimary productivity production at the plume/oceanic interface of the Mi ssissippi River. Continental Shelf Research 10:639-664 Macko SA, Engle MH, Parker PL (1993) Early diagenes is of organic matter in sediments. In: Engle MH, Macko SA (eds) Organic Geo chemistry. Plenum Press, New York, p 211-223
82 McIsaac GF, David MB, Gertner GZ, Goolsby DA (2001) Nitrate flux in the Mississippi River. Nature 414:166-167 McIsaac GF, Hu X (2004) Net N input and riverine N export from Illinois agricultural watersheds with and without extensive tile drainage. Biogeochemistry 70:251-271 McKee B, Baskaran M (1999) Sedimentary processes of Gulf of Mexico estuaries. In: Bianchi TS, Pennock JR, Twilley RR ( eds) Biogeochemistry of Gulf of Mexico Estuaries Mehnert E, Hwang H-H, Johnson TM, Sanford RA, Beaum ont WC, Holm TR (2007) Denitrification in the shallow ground water of a tile-drained, agricultural watershed. Journal of Environmental Qu ality 36:80-90 Meyers PA (1994) Preservation of elemental and isot opic source identification of sedimentary organic matter. Chemical Geology 114:28 9-302 Meyers PA (1997) Organic geochemical proxies of pal eoceanographic, paleolimnologic, and paleoclimatic processes. Organ ic Geochemistry 27:213-250 Nelsen TA, Blackwelder P, Hood T, McKee BA, Romer N Alvarez-Zarikian C, Metz S (1994) Time-based correlation of biogenic, l ithogenic and authigenic sediment components with anthropogenic i nputs in the Gulf of Mexico NECOP study area. Estuaries 17:873-885 Nixon SW (1995) Coastal marine eutrophication: a de finition, social causes, and future concerns. Ophelia 41:199-219 O'Reilly CM, Hecky RE (2002) Interpreting stable is otopes in food webs: Recognizing the role of time averaging at different trophic levels. Limnology and Oceanography 47:306-309 Osterman LE (2003) Benthic foraminifers from the co ntinental shelf and slope of the Gulf of Mexico: an indicator of shelf hypoxia. Estuarine, Coastal and Shelf Science 58:17-35 Osterman LE, Poore RZ, Castenson ED (2000) Benthic foraminiferal census data and radiocarbon dates from the Gulf of Mexico (Mata gorda 1A core), USGS Open-File Report 2000-0211 Osterman LE, Poore RZ, Swarzenski PW (2008) The las t 1000 years of natural and anthropogenic low-oxygen bottom-water on the Lo uisiana shelf, Gulf of Mexico. Marine Micropaleontology 66:291-303
83 Osterman LE, Poore RZ, Swarzenski PW, Turner RE (20 05) Reconstructing a 180 yr record of natural and anthropogenic induced low-oxygen conditions from Louisiana continental shelf sediments. Geology 33:329-332 Pancost RD, Pagani M (2006) Controls on the carbon isotopic compositions of lipids in marine environments. In: Volkman JK (ed) Marine Organic Matter: Biomarkers, Isotopes and DNA, Vol 2. Springer, Berl in, p 209-249 Panno SV, Hackley KC, Kelly WR, Hwang H-H (2006) Is otopic evidence of nitrate sources and denitrification in the Mississi ppi River, Illinois. Journal of Environmental Quality 35:495-504 Peterson BJ, Fry B (1987) Stable isotopes in ecosys tem studies. Annual Review of Ecological Systems 18:293-320 Platon E, Sen Gupta BK, Rabalais NN, Turner RE (200 5) Effect of seasonal hypoxia on the benthic foraminiferal community of t he Louisiana inner continental shelf: The 20th century record. Marine Micropaleontology 54:263-283 Quiones-Rivera ZJ, Wissel B, Justi D, Fry B (2007) Partitioning oxygen sources and sinks in a stratified, eutrophic coasta l ecosystem using stable oxygen isotopes. Marine Ecology Progress Series 342 :69-83 Rabalais NN, Atilla N, Normandeau C, Turner RE (200 4) Ecosystem history of Mississippi River-influenced continental shelf reve aled through preserved phytoplankton pigments. Marine Pollution Bulletin 4 9:537-547 Rabalais NN, Turner RE (eds) (2001a) Coastal Hypoxi a: Consequences for Living Resources and Ecosystems, Coastal and Estuar ine Studies Vol 58. American Geophysical Union, Washington, D.C. Rabalais NN, Turner RE (2001b) Hypoxia in the north ern Gulf of Mexico: description, causes and change. In: Rabalais NN, Tu rner RE (eds) Coastal Hypoxia: Consequences for Living Resources and Ecosystems. Coastal and Estuarine Studies Vol. 58, American Geo physical Union, Washington, D.C., p 1-36 Rabalais NN, Turner RE, Dortch Q, Justi D, Bierman Jr. VJ, Wiseman Jr. WJ (2002a) Nutrient-enhanced productivity in the north ern Gulf of Mexico: past, present and future. Hydrobiologia 475/476:3963 Rabalais NN, Turner RE, Justi D, Dortch Q, Wiseman Jr. WJ (1999) Characterization of Hypoxia: Topic 1 Report for the Integrated Assessment of Hypoxia in the Gulf of Mexico, NOAA C oastal Ocean Program, Silver Spring, Maryland
84 Rabalais NN, Turner RE, Justi D, Dortch Q, Wiseman Jr. WJ, Sen Gupta BK (1996) Nutrient changes in the Mississippi River an d system responses on the adjacent continental shelf. Estuaries 19:386-40 7 Rabalais NN, Turner RE, Scavia D (2002b) Beyond sci ence into policy: Gulf of Mexico hypoxia and the Mississippi River. BioScienc e 52:129-142 Rabalais NN, Turner RE, Sen Gupta BK, Boesch DF, Ch apman P, Murrell MC (2007a) Hypoxia in the northern Gulf of Mexico: doe s the science support the plan to reduce, mitigate, and control hypoxia? Estuaries and Coasts 30:753-772 Rabalais NN, Turner RE, Sen Gupta BK, Platon E, Par sons ML (2007b) Sediments tell the history of eutrophication and hy poxia in the northern Gulf of Mexico. Ecological Applications 17:S129-S14 3 Rabalais NN, Turner RE, Wiseman Jr. WJ (2001) Hypox ia in the Gulf of Mexico. Journal of Environmental Quality 30:320-329 Rabalais NN, Turner RE, Wiseman Jr. WJ (2002c) Gulf of Mexico hypoxia, a.k.a. "The Dead Zone". Annual Review of Ecology and Syste matics 33:235-263 Rabalais NN, Turner RE, Wiseman Jr. WJ, Dortch Q (1 998) Consequences of the 1993 Mississippi River flood in the Gulf of Mex ico. Regulated Rivers: Research & Management 14:161-177 Rabalais NN, Wiseman Jr. WJ, Turner RE (1994) Compa rison of continuous records of near-bottom dissolved oxygen from the hy poxia zone along the Louisiana Coast. Estuaries 17:850-861 Rowe GT, Kaegi MEC, Morse JW, Boland GS, Briones EG E (2002) Sediment community metabolism associated with continental sh elf hypoxia, northern Gulf of Mexico. Estuaries 25:1097-1106 Sackett WM (1964) The depositional history and isot opic organic carbon composition of marine sediments. Marine Geology 2:1 73-185 Savage C (2005) Tracing the influence of sewage nit rogen in a coastal ecosystem using stable nitrogen isotopes. Ambio 34: 145-150 Savage C, Elmgren R (2004) Macroalgal (Fucus vesicu losus) d15N values trace decrease in sewage influence. Ecological Applicatio ns 14:517-526 Scavia D, Rabalais NN, Turner RE, Justi D, Wiseman Jr. WJ (2003) Predicting the response of Gulf of Mexico hypoxia to variation s in Mississippi River nitrogen load. Limnology and Oceanography 48:951-95 6
85 Seitzinger S, Harrison JA, Bohlke JK, Bouwman AF, L owrance R, Peterson B, Tobias C, Van Drecht G (2006) Denitrification acros s landscapes and waterscapes: A synthesis. Ecological Applications 1 6:2064-2090 Sen Gupta BK, Turner RE, Rabalais NN (1996) Seasona l oxygen depletion in continental-shelf waters of Louisiana: historical r ecord of benthic foraminifers Geology 24:227-230 Swarzenski PW, Campbell PL, Osterman LE, Poore RZ ( 2008) A 1000-year sediment record of recurring hypoxia off the Missis sippi River: The potential role of terrestrially-derived organic mat ter inputs. Marine Chemistry 109:130-142 Trefry JH, Metz S, Nelsen TA, Trocine RP, Eadie BJ (1994) Transport of particulate organic carbon by the Mississippi River and its fate in the Gulf of Mexico. Estuaries 17:839-849 Turner RE, Qureshi N, Rabalais NN, Dortch Q, Justi D, Shaw RF, Cope J (1998) Fluctuating silicate:nitrate ratios and coas tal plankton food webs. Proceedings of the National Academy of Sciences, US A 95:13048-13051 Turner RE, Rabalais NN (1991) Changes in Mississipp i River water quality this century. BioScience 41:140-147 Turner RE, Rabalais NN (1994) Coastal eutrophicatio n near the Mississippi river delta. Nature 368:619-621 Turner RE, Rabalais NN (2003) Linking landscape and water quality in the Mississippi River Basin for 200 years. BioScience 5 3:563-572 Turner RE, Rabalais NN, Alexander RB, McIsaac GF, H owarth RW (2007) Characterization of nutrient, organic carbon, and s ediment loads and concentrations from the Mississippi River into the northern Gulf of Mexico. Estuaries and Coasts 30:773-790 Turner RE, Rabalais NN, Justi D (2006) Predicting summer hypoxia in the northern Gulf of Mexico: Riverine N, P, and Si load ing. Marine Pollution Bulletin 52:139-148 Turner RE, Rabalais NN, Justi D (2008) Gulf of Mexico hypoxia: Alternate states and a legacy. Environmental Science & Technology Wada E, Hattori A (1991) Nitrogen in the sea: Forms abundances, and rate processes. CRC
86 Walker ND, Rabalais NN (2006) Relationships among s atellite chlorophyll a river inputs, and hypoxia on the Louisiana continental sh elf, Gulf of Mexico. Estuaries and Coasts 29:1081-1093 Wiseman Jr. WJ, Rabalais NN, Turner RE, Dinnel SP, MacNaughton A (1997) Seasonal and interannual variability within the Lou isiana Coastal Current: stratification and hypoxia. Journal of Marine Syste ms 12:237-248 Wissel B, Fry B (2005) Sources of particulate organ ic matter in the Mississippi River, USA. Large Rivers 15:105-118 Zimmerman AR, Canuel EA (2000) A geochemical record of eutrophication and anoxia in Chesapeake Bay sediments: anthropogenic i nfluence on organic matter composition. Marine Chemistry 69:117-137 Zimmerman AR, Canuel EA (2002) Sediment geochemical records of eutrophication in the mesohaline Chesapeake Bay. Li mnology and Oceanography 47:1084-1093
87 Appendix A: Geochemistry Data
88 Appendix A: Geochemistry data for the three cores used in this study. Sample age is given when available. For completeness, PEB index values have been included, but these values were generated by Lisa Osterman at the USGS and have been published previo usly. PE0305-GC1 Age (Years) Sample Depth (cm) Organic Carbon (%) 15N 13C PEB C:N Molar 2002.0 0 36.19 2000.5 1 29.95 1997.6 2 28.24 1995.2 3 -22.21 23.79 1991.7 4 0.981 4.944 -23.24 29.17 1988.8 5 0.866 5.861 -22.19 29.19 22.84 1985.8 6 0.829 -21.91 19.26 20.38 1982.9 7 0.880 22.46 25.83 1979.9 8 0.808 4.844 22.58 25.80 1977.0 9 0.843 4.841 21.15 26.24 1974.1 10 0.890 -23.25 24.49 21.56 1971.1 11 0.803 4.329 23.19 33.93 1969.1 12 0.823 4.234 -21.07 17.08 23.29 1965.2 13 0.837 3.442 -23.48 17.37 28.13 1962.3 14 0.799 -22.77 14.41 15.58 1959.4 15 0.712 9.25 27.01 1956.4 16 0.957 3.481 -22.32 14.71 24.22 1953.5 17 0.768 4.020 -24.63 12.63 27.24 1950.5 18 0.771 4.719 -24.38 13.04 27.71 1947.6 19 0.812 -22.54 9.83 19.44 1944.6 20 0.711 16.30 1941.7 21 0.728 -20.02 15.53 1938.8 22 0.803 4.273 14.04 24.75 1935.8 23 0.756 18.50 27.82 1932.9 24 3.650 14.33 27.97 1929.9 25 -27.69 16.22 20.62 1927.0 26 0.764 4.124 -24.07 14.20 30.60 1924.1 27 0.680 3.800 -23.10 9.16 26.84 1921.1 28 0.679 3.184 10.72 1918.2 29 0.621 -24.90 11.04 20.06 1915.2 30 0.712 5.52 26.20
89 Appendix A: (continued) Age (Years) Sample Depth (cm) Organic Carbon (%) 15N 13C PEB C:N Molar 1912.3 31 0.679 6.10 36.13 1909.4 32 0.600 4.152 -23.46 9.26 26.81 1906.4 33 0.650 3.185 -24.20 10.20 26.44 1903.5 34 0.616 3.996 -25.96 8.96 32.52 1900.5 35 0.714 2.677 -25.40 18.29 40.67 1897.6 36 0.598 3.572 -24.07 19.01 1894.6 37 0.590 3.887 -22.07 34.72 1891.7 38 0.623 3.534 -22.95 8.03 27.60 1888.8 39 0.752 8.30 26.53 1885.8 40 0.735 3.454 -23.50 5.62 26.99 41 0.611 -25.40 16.31 26.04 42 0.601 -24.85 13.77 20.93 43 0.610 3.761 -24.40 13.77 26.81 44 0.666 3.290 -23.49 14.80 45 0.592 2.567 -24.89 10.28 27.27 46 0.664 3.687 13.24 26.19 47 0.608 9.48 28.32 48 0.667 13.76 21.14 49 0.685 15.73 25.31 50 0.667 13.36 28.72 51 0.581 -23.34 18.88 26.07 52 0.589 3.460 -25.15 19.69 53 0.703 3.017 -24.57 11.95 26.23 54 0.642 -25.95 12.34 21.55 55 0.676 10.53 24.61 56 0.678 -24.77 15.48 22.40 57 0.622 13.81 21.85 58 1.940 -23.01 13.51 23.35 59 0.611 3.884 12.84 30.13 60 0.626 3.816 18.95 61 0.629 3.857 13.20 43.68 62 0.640 2.930 -26.32 10.14 29.02 63 0.643 15.50 30.83 64 0.643 3.671 8.37 25.40 65 0.663 7.16 24.77 66 0.615 -24.32 6.78 22.80
90 Appendix A (continued) Age (Years) Sample Depth (cm) Organic Carbon (%) 15N 13C PEB C:N Molar 67 0.679 5.64 33.23 68 0.632 -25.80 7.63 69 0.606 3.261 -28.07 8.41 31.04 70 0.651 3.680 -25.53 8.63 28.79 71 0.689 5.22 26.49 72 0.605 3.937 8.70 28.78 73 0.615 3.728 -24.86 7.80 27.26 74 0.607 2.902 -24.17 12.96 34.04 75 0.651 3.180 -23.04 14.77 28.11 77 0.615 3.847 -23.95 11.03 29.20 78 0.625 3.939 -23.88 9.88 24.45 79 0.596 7.40 28.27 80 0.664 3.315 -25.36 6.76 34.26 81 0.634 3.649 -25.22 3.93 30.05 82 0.666 -26.45 4.06 23.89 83 0.642 3.590 -22.73 6.77 30.47 84 0.601 3.952 -25.28 17.17 85 0.583 3.950 -24.49 11.97 34.11 86 0.616 3.638 8.30 87 0.590 13.23 39.57 88 0.591 3.592 12.80 39.88 89 0.625 3.300 -24.96 15.49 28.70 90 0.695 4.670 -23.38 12.06 26.51 91 0.656 3.510 -24.37 15.69 31.82 92 0.753 4.115 -23.02 6.42 93 0.774 2.690 -25.07 4.26 33.07 94 0.763 3.320 -24.93 6.67 34.85 95 0.730 13.71 35.64 96 0.762 3.480 -24.97 17.78 31.88 97 0.841 3.609 -25.66 22.97 42.92 98 0.522 3.923 -25.00 6.85 33.17 99 0.668 1.990 -26.42 43.95 100 0.666 3.579 -23.36 101 0.731 3.501 -24.44 28.72 102 0.742 3.916 -23.78 28.22 103 0.669 30.20
91 Appendix A (continued) Age (Years) Sample Depth (cm) Organic Carbon (%) 15N 13C PEB C:N Molar 104 0.645 2.760 -22.42 21.92 28.99 105 0.672 2.967 -24.69 17.55 31.55 106 0.653 2.810 -24.22 32.38 31.86 107 0.626 3.495 -22.27 8.71 29.14 108 0.706 3.653 -22.62 10.91 109 0.653 3.365 7.11 35.65 110 0.733 3.830 -22.76 4.58 25.33 111 0.703 8.21 27.83 112 0.699 3.760 -20.79 4.55 24.06 113 0.573 3.478 -27.05 2.82 45.04 114 0.712 3.060 -21.82 2.48 27.32 115 0.683 3.780 -21.38 6.15 30.01 116 0.685 3.873 4.27 118 0.658 3.796 -22.63 5.61 28.09 119 0.657 5.91 27.62 120 0.708 2.54 31.95 121 0.635 4.282 11.98 31.09 122 0.703 3.310 -22.90 6.23 29.69 123 0.715 3.630 -21.50 11.59 30.78 124 0.710 3.833 5.11 125 0.709 3.218 -21.97 10.96 30.99 126 0.673 3.740 -22.89 11.43 29.24 127 0.707 6.38 30.31 128 0.754 3.370 -21.37 11.48 32.03 129 0.836 3.094 -22.63 6.47 31.24 130 0.875 3.060 58.55 27.85 131 1.029 3.260 -21.91 62.00 39.75 132 0.812 3.763 13.87 133 0.641 7.72 43.14 134 0.597 3.980 -25.75 10.47 34.87 135 0.680 3.832 -23.58 7.56 28.83 136 0.700 2.540 -21.84 10.50 31.97 137 0.772 3.086 -22.88 4.00 33.39 138 0.807 3.478 -23.00 13.94 32.34 139 0.680 14.29 37.98 140 0.653 3.544 16.40 141 0.712 2.624 15.82 33.70
92 Appendix A (continued) Age (Years) Sample Depth (cm) Organic Carbon (%) 15N 13C PEB C:N Molar 142 0.732 3.890 -24.15 21.55 27.54 143 0.699 3.540 -24.63 25.63 29.85 144 0.793 3.454 -24.54 28.93 145 3.153 -24.89 22.17 47.15 146 0.483 3.680 -23.91 21.35 25.15 147 3.818 -22.74 25.26 34.76 148 3.066 -22.69 18.67 24.48 149 0.583 3.040 -23.51 22.15 26.47 150 4.720 -24.15 17.31 29.51 151 0.369 4.200 -24.20 16.42 152 4.821 -23.03 14.07 21.80 153 0.502 13.04 154 2.948 -23.18 17.40 25.25 155 0.482 17.95 41.28 156 3.090 -22.58 11.04 25.69 157 0.437 3.613 -23.20 17.16 22.92 159 0.454 3.752 16.73 26.67 160 3.880 -22.84 8.41 22.46 161 5.502 -24.07 14.79 30.88 162 0.532 4.453 -23.15 13.93 22.78 163 0.512 11.11 25.70 164 4.065 -22.50 18.24 19.02 PE0305-BC1 Age Sample Depth (cm) Organic Carbon (%) 13C 15N PEB C:N Molar 2000.5 0-1 1.560 -21.826 5.85 27.74 1997.6 1-2 1.431 -21.633 5.96 32.21 1994.6 2-3 1.651 -21.374 5.59 31.84 8.47 1991.7 3-4 1.577 -22.082 5.63 31.83
93 Appendix A (continued) Age (Years) Sample Depth (cm) Organic Carbon (%) 15N 13C PEB C:N Molar 1988.8 4-5 1.852 -22.806 5.92 29.23 8.76 1985.8 5-6 1.789 -22.118 5.77 22.61 8.97 1982.9 6-7 1.961 -21.963 5.99 31.44 8.69 1979.9 7-8 1.592 -22.132 5.77 29.56 1977.0 8-9 1.453 -22.163 6.03 30 9.29 1974.1 9-10 1.339 -22.254 5.88 26.7 1971.1 10-11 1.554 -21.995 5.97 18.83 1968.2 11-12 1.787 -21.932 5.18 21.53 1965.2 12-13 1.432 -23.940 4.97 20.52 9.14 1962.3 13-14 1.557 -21.993 5.00 23.88 1959.4 14-15 1.511 -22.043 5.32 20.48 8.81 1956.4 15-16 1.622 -21.954 4.99 18.69 1953.5 16-17 1.425 -21.751 4.84 15.86 8.55 1950.5 17-18 1.457 -22.241 4.22 1947.6 18-19 1.791 -22.577 2.95 12.3 1941.7 20-21 1.491 -23.376 3.04 15.54 9.36 1938.8 21-22 1.351 -23.274 3.16 12.33 1935.8 22-23 1.241 -23.619 3.20 8.82 1932.9 23-24 1.402 -23.215 3.14 8.22 1929.9 24-25 1.166 -23.703 3.32 16.08 1927.0 25-26 1.135 -23.701 3.61 15.77 1924.1 26-27 1.173 -24.015 3.73 12.01 1921.1 27-28 1.278 -23.872 3.43 10.84 1918.2 28-29 1.419 -23.864 4.16 12.89 9.08 1915.2 29-30 1.379 -23.661 3.94 17.28 1912.9 30-31 1.303 -23.741 4.08 15.2
94 Appendix A (continued) MRD05-BC6 Age Sample Depth (cm) Organic Carbon (%) 13C 15N PEB C:N Molar 2002.9 0-1 1.104 14.65 1998.6 1-2 0.957 n 22.06 9.70 1994.3 2-3 1.009 nn 17.35 9.00 1990.0 3-4 0.916 17.93 8.81 1985.7 4-5 0.961 15.15 10.08 1981.4 5-6 0.948 nn 15.27 8.98 1977.1 6-7 0.955 r 21.51 9.23 1972.8 7-8 0.994 20.47 9.20 1968.5 8-9 0.977 r 22.87 8.77 1964.3 9-10 0.978 n 15.71 9.41 1960.0 10-11 0.795 nr 16.61 9.59 1955.7 11-12 0.781 16.18 8.85 1951.4 12-13 0.736 nn 15.78 9.11 1947.1 13-14 0.628 17.43 9.56 1942.8 14-15 0.879 nn 13.21 9.94 1938.5 15-16 0.908 n 11.92 9.72 1929.9 17-18 0.743 nrr 12.46 10.32 1925.7 18-19 0.664 5.75 9.92 1921.4 19-20 0.800 n 9.04 9.28 1917.1 20-21 0.679 r 9.64 9.70 1912.8 21-22 0.901 7.01 9.52 1908.5 22-23 0.495 5.23 9.56 1904.2 23-24 0.971 6.42 9.91 1899.9 24-25 0.784 9.71 9.75 1895.6 25-26 0.995 nn 8.04 9.45 1891.3 26-27 0.799 n 6.43 9.77 1887.0 27-28 0.941 n 9.07 9.22 1882.8 28-29 0.742 n 9.65 9.30 1878.5 29-30 0.920 8.05 9.98 1874.2 30-31 0.663 9.61 9.39 1869.9 31-32 0.857 rr 10.22 1865.6 32-33 0.740 rr 9.90 1861.3 33-34 0.721 9.70 1857.0 34-35 0.708 n 5.45 9.84
95 Appendix A (continued) Age (Years) Sample Depth (cm) Organic Carbon (%) 13C 15N PEB C:N Molar 1852.7 35-36 0.808 rr 13.40 9.38 1848.4 36-37 0.694 r 6.34 9.26 1844.2 37-38 0.857 6.11 9.70 1839.9 38-39 0.708 12.73 9.58 1835.6 39-40 0.840 nr 6.02 9.95 1831.3 40-41 0.678 r 6.54 9.71 1827.0 41-42 0.733 8.37 9.83