DRC-2010-006004 - 0901a068801dcfd4

DRC-2010-006004
DENISO
MINES
November 15, 2010
Denison Mines (USA) Corp.
1050 17th street, Suite 950
Denver, CO 80265
USA
Tel: [phone]
Fax : [phone]
www.denlsonmines.com
VIA PDF AND FEDERAL EXPRESS
Mr. Rusty Lundberg
Co-Executive Secretary
Utah Water Quality Board
State of Utah Department of Environmental Quality
195 North 1950 West
P.O. Box 144850
Salt Lake City, UT 84114-4850
Dear Mr. Lundberg:
Re: Stipulated Consent Agreement UGW09-03 dated January 27, 2009; Nitrate Contamination
Investigation Report, White Mesa Uranium Mill Site, Blanding Utah dated December 30, 2009; and
letter dated October 5, 2010 from the Executive Secretary
Reference is made to the Stipulated Consent Agreement UGW09-03 dated Januaiv 27, 2009 (the "Consent
Agreement') related to nitrate contamination at Denison Mines (USA) Corp's ("Denison's") White Mesa Uranium
Mill Site, near Blanding Utah (the "Mill"), and to the Nitrate Contamination Investigation Report, White Mesa
Uranium Mill Site, Blanding Utah, dated December 30, 2009 (the "CIR") submitted to the Utah Division of
Radiation Control ("DRC") pursuant to Item 6.A of the Consent Agreement.
Reference is also made to the letter dated October 5, 2010 from the Executive Secretary of the Utah Radiation
Control Board/Co-Executive Secretary of the Utah Water Quality Board (the "Executive Secretary") notifying
Denison of his determination that the CIR has not fully satisfied the Consent Agreement requirements (the
"DRC Notice").
The purpose of this letter is to summarize the additional studies identified to date that may be addressed
in a revised CIR, including the additional studies suggested in the DRC Notice, any proposed additional
studies relating to the new information introduced by Denison at a meeting held on October 26, 2010
between Denison and DRC staff (the "October 26 Meeting"), and any other additional studies that
Denison believes may be relevant. This letter also proposes a date for a meeting between DRC Staff and
Denison technical and regulatory staff to discuss the foregoing studies and any associated matters, to
agree on the studies to be performed and the manner of performing those studies, and to develop a plan
and schedule for performing such studies and for submittal of a revised CIR.
1) Additional Studies Suggested in the DRC Notice
In the DRC Notice, DRC suggested a number of additional studies that could be performed in support of a
revised CIR. DRC's observations and suggestions are discussed below.
OENISOJ)~~
MINES
November 15,2010
VIA PDF AND FEDERAL EXPRESS
Mr. Rusty Lundberg
Co-Executive Secretary
Utah Water Quality Board
State of Utah Department of Environmental Quality
195 North 1950 West
P.O. Box 144850
Salt Lake City, UT 84114-4850
Dear Mr. Lundberg:
Denison Mines (USA) Corp.
1050 17th Street, Suite 950
Denver, CO 80265
USA
Tel: 303 628·7798
Fax: 303 389·4125
www.denisonmines.com
Re: Stipulated Consent Agreement UGW09-03 dated January 27, 2009; Nitrate Contamination
Investigation Report, White Mesa Uranium Mill Site, Blanding Utah dated December 30, 2009; and
letter dated October 5,2010 from the Executive Secretary
Reference is made to the Stipulated Consent Agreement UGW09-03 dated January 27, 2009 (the "Consent
Agreement") related to nitrate contamination at Denison Mines (USA) Corp's ("Denison's") White Mesa Uranium
Mill Site, near Blanding Utah (the "Mill"), and to the Nitrate Contamination Investigation Report, White Mesa
Uranium Mill Site, Blanding Utah, dated December 30, 2009 (the "CIR") submitted to the Utah Division of
Radiation Control ("DRC") pursuant to Item 6.A of the Consent Agreement.
Reference is also made to the letter dated October 5, 2010 from the Executive Secretary of the Utah Radiation
Control Board/Co-Executive Secretary of the Utah Water Quality Board (the "Executive Secretary") notifying
Denison of his determination that the CIR has not fully satisfied the Consent Agreement requirements (the
"DRC Notice").
The purpose of this letter is to summarize the additional studies identified to date that may be addressed
in a revised CIR, including the additional studies suggested in the DRC Notice, any proposed additional
stUdies relating to the new information introduced by Denison at a meeting held on October 26, 2010
between Denison and DRC staff (the "October 26 Meeting"), and any other additional studies that
Denison believes may be relevant. This letter also proposes a date for a meeting between DRC Staff and
Denison technical and regulatory staff to discuss the foregoing studies and any associated matters, to
agree on the studies to be performed and the manner of performing those studies, and to develop a plan
and schedule for performing such studies and for submittal of a revised CIR.
1) Additional Studies Suggested in the DRC Notice
In the DRC Notice, DRC suggested a number of additional studies that could be performed in support of a
revised CIR. DRC's observations and suggestions are discussed below.
a) Nitrate Source Determination (Frog Pond)
DRC concluded that Denison's conclusion in the CIR that the source of the nitrate/chloride
contamination was likely from use of Frog Pond water is "unsubstantiated with direct and reliable
evidence" and suggested a number of additional studies that could be performed in order to
substantiate this conclusion, including the following:
i) Blanding City Wastewater Discharge to the Frog Pond
DRC concluded that Denison has offered no quantifiable data to support any contamination from
the City Lagoon to the Frog Pond, and that additional information and sources to support these
claims are required in order to substantiate these claims.
iij Demonstration of Slug Flow Behavior
DRC concluded that "no clear slug behavior is evident to the DRC, as claimed in the CIR, since
Nitrate and Chloride concentrations downgradient from the Mill are within the range of upgradient
concentrations. The statements by DUSA regarding slug flow behavior need further explanation
by DUSA, and need to be justified by sufficient data and analysis"
DRC proposed that such additional work should include but not be limited to the following
additional studies:
• Additional wells and borings around the Frog Pond;
• Hydraulic evaluation including equipotential data utilizing additional testing around the
Frog Pond;
• Characterization of any ground water mounding potentially caused by the Frog Pond
• Geologic Information;
• An evaluation of the elevation of the Brushy Basin shale upper contact in the area of
Coral Canyon and an interpretation of the hydraulic conditions between the Frog Pond
and the Mill Site;
• Additional Nitrate + Nitrite (as N) and Chloride shallow ground water quality data for
new upgradient wells installed near or around the Frog Pond area; and
• Additional characterization to determine the presence and location of a preferential
ground water flow path that allowed a 30 year or less nitrate travel time from the Frog
Pond to the Mill Site.
iiij Hydraulic Gradient and Flow Path Issues
DRC concluded that Denison has not offered analysis of potential ground water hydraulic
paths from the City Lagoon to the Frog Pond (e.g ., has not mapped local stream tubes),
and that the arguments in the CIR based on flow velocities from upgradient sources to the
current plume location were "inconclusive and indicated that assumptions of preferential
flow paths and heterogeneities would be needed to explain the observed plume location,
however, the claims of preferential flow paths were not studied or supported." DRC did not
recommend any particular studies that could be performed.
b) Need for Additional Source (Potential Multiple Source) Investigation
DRC noted that the CIR provided some "limited justification" that certain sources identified
in the Source Review Report contained in the CIR could not be the sole sources of the
plume concentrations based on low contributions of nitrate, time of use, etc. However,
DRC concluded that it is not appropriate to negate these sources (e.g., the SAG leach
field) without additional assessment as recommended in the Source Review Report. DRC
noted that such additional assessment should include the following:
OENISOJ)~~
MINES
2
• In-situ soil sampling; and
• Shallow ground water sampling.
c) Potential Option for Additional Study/Justification for an "Upgradient" Source
DRC also noted that, in order to distinguish an "upgradient source", Denison may opt to perform
isotopic analysis to provide information regarding an "upgradient" contaminant signature (isotopic
fingerprint) to confirm that the same signature is present at multiple locations and depths both
beneath the Mill and at "upgradient" locations. DRC also noted that the study should also include
an evaluation of "downgradient" nitrate concentrations at multiple locations and depths.
2) Additional Studies Proposed by Denison
After reviewing DRC's recommendations and suggestions in the DRC Notice, and considering other
possible studies identified by Denison, Denison proposes that the following additional studies be
undertaken to form the basis for a revised CIR.
a) Additional Studies Relating to Frog Pond as a Potential Source
It is clear from the DRC Notice that DRC does not believe the nitrate/chloride contamination at
the Mill site could have originated from the Frog Pond.
Denison's conclusions in the CIR relating to the Frog Pond and the City Lagoons are based on
anecdotal evidence and on assumptions that historical practices at the City Lagoons may have
resembled practices followed by other similar facilities across the country in the past, which
practices have resulted in nitrate and chloride contamination to groundwater at a number of those
facilities. However, as has been observed at such other facilities, groundwater contamination
may have been caused by historical practices that are no longer being followed, and it is
therefore very difficult to determine the source of the contamination, based on currently available
information. This fact, combined with the fact that Denison does not control the land near the
Lagoons and the Frog Pond and cannot mandate that the current owners participate in any field
investigations, has resulted in Denison not recommending the pursuit of the additional studies
recommended in Section 1 (a) above at this time.
Denison believes that information relevant to the potential sources of contamination that may be
identified in the studies discussed below may provide additional evidence that could warrant
pursuit of additional investigations relating to the Frog Pond in the future. However, at this time,
Denison believes it is better to direct its resources at some of the other avenues discussed below.
b) Additional Source (Potential Multiple Source) Investigation
i) Geoprobe study
Denison proposes to perform a geoprobe study of unconsolidated soils in areas at the site
that could have had a large enough potential nitrate or chloride source and a large enough
hydraulic head to potentially have reached the perched groundwater zone at the site, as
identified in the Source Review Report that was included in the CIR. Samples will be
collected at 6 to 10 foot depths from such specific potential sources and analyzed by the
synthetic precipitation leaching procedure (SPLP) (EPA Standard Method 1312) for nitrate
and chloride. Denison will propose to DRC a plan and schedule for carrying out this study,
including the potential areas to be included in the study.
OENISONI)~~
MINES
3
ii) Shallow Groundwater Sampling
Denison believes that the nitrate and chloride plumes are sufficiently characterized from the
existing monitoring wells placed in the perched aquifer. Therefore, Denison is not proposing
that any additional wells be drilled at this time. However, the isotopic sampling and analysis
discussed in paragraph (d) below will be performed at selected existing monitoring wells at
the site, which may provide additional useful shallow groundwater sampling information.
iii) Mass Balance Analysis
Denison also proposes that it would be beneficial to carry out a mass balance analysis. Such
an analysis will analyze the concentrations of nitrate and chloride in the groundwater, the
hydrogeologic features associated with the plume, any mounding of water in the perched
zone or lack thereof and other relevant factors, and determine the concentrations or amounts
of nitrate and chloride that would be necessary in order to generate the concentrations
identified in the plume. Denison believes that such an analysis will provide useful information
in eliminating certain potential sources from further consideration and could possibly identify
new potential sources.
ivY Supplemental Historic Source Study
Denison proposes to extend the existing Source Review Report to address or further
evaluate existing and new potential sources, such as the military installation/ordinance
source that was located near the Mill, and any overlooked ranching sources.
c) Investigation of Potential Natural Reservoir of Nitrate and Chloride in Solis Beneath the Mill
Site
At the October 26 Meeting, Denison mentioned the possibility that the nitrate and chloride
contamination at the Mill site and in surrounding areas may be caused by a large reservoir of
bioavailable nitrogen and chloride that has been observed to have been accumulating in
subsoil zones of arid regions over time. Attached to this letter is an article entitled A
Reservoir of Nitrate Beneath Desert Soils, dated November 7, 2003, written by Walvoord et
ai, that addresses this concept. In that article, the authors note that "[r]ecent studies indicate
that subsoil N03 reservoirs are readily mobilized to groundwater when desert land is
converted to irrigation ... " and that "[d]am construction or changes in climate and vegetation
could likewise mobilize subsoil nitrate reservoirs, with local to regional effects."
Denison was not aware of this potential source of the nitrate and chloride contamination at
the time it submitted the CIR to the Executive Secretary.
Denison proposes to undertake a coring study to determine if such a phenomenon exists at
and around the Mill site that could have caused or contributed to the nitrate and chloride
contamination under investigation. This study would test the possibility that a reservoir of
nitrate and chloride exists at some level in the geologic column above groundwater due to
lack of distributed recharge to groundwater over an extended period of time.
Core would be taken from an area that has not been affected by focused recharge such as
the wildlife ponds or intermittent streams and drainage channels. Samples of core would be
taken at regular intervals, moisture content measured, and leached with specific amounts of
double distilled water to determine the concentrations of nitrate and chloride present in soil
moisture. Mass balance calculations would then integrate the mass of nitrate and chloride in
soil moisture to determine if the total mass is sufficient to account for the observed
concentrations in groundwater. Denison will propose to DRC a plan and schedule for
carrying out this investigation.
OENISOJ)~~
MINES
4
d) Additional Study/Justification for an "Upgradient" Source
The DRC Notice suggests that studies be aimed at identifying possible "upgradient" sources.
However, it should be noted that any additional studies should also be aimed at identifying
sources not related to Mill activities, which need not be upgradient sources.
Denison agrees that stable isotopic analysis may be employed to distinguish different
sources of nitrate contamination that may distinguish one or more sources that are not
related to Mill activities.
Denison proposes to perform such an analysis at the Mill site to identify the isotopic
fingerprint of the nitrate contamination. This may provide information that will be useful in
determining if the nitrate contamination is due to one or more sources that are not related to
Mill activities. This study will include an evaluation of upgradient and far downgradient
contamination.
Denison will propose to the Executive Secretary a plan and schedule for carrying out this
investigation.
3) Suggested Date for Meeting
Denison proposes that DRC staff meet with Denison technical and regulatory staff on November 30,2010
to discuss the additional studies Denison proposes to undertake, as summarized above, and any other
investigations DRC and Denison may identify. If that date does not work, Denison would be pleased to
accommodate DRC's schedule.
It is hoped that at the meeting DRC and Denison will agree on a proposed course of action in general
terms, including the objectives of each study and the criteria that must be satisfied in order for DRC to
consider the revised CIR to meet all applicable requirements. Denison will, within two weeks of the
meeting, submit detailed plans and schedules for each additional study to be undertaken, for Executive
Secretary approval.
If you have any questions or require any further information, please contact the undersigned.
Yours very truly,
DENIS MIN
D I j F denlund
Vic Presid nt, Regulatory Affairs and Counsel
cc: Ron F. Hochstein
Harold R. Roberts
Jo Ann S. Tischler
David E. Turk
Daniel W. Erskine, Ph.D. INTERA, Inc.
5
OENISOJ) ~
MINES
Yungay area, the nature of the oxidant remains
unexplained. Photochemical reactions initiated
by sunlight continually produce oxidants in the
lower atmosphere and surface. However, in
most soils, biological production of reduced
organic material completely dominates the net
redox state of soils. When biological production
is less than the photochemical production of
oxidation, then the soil will become oxidizing.
The transition from biologically dominated
soils to photochemically dominated soils
appears to be abrupt. Whichever process
dominates will shift the redox state in one
direction or another. In the Atacama, there
is a gradual decline in biological activity as
conditions became drier, yet near the ex-
treme arid region there is an abrupt transi-
tion to very low bacterial levels and low
organic content.
It is unlikely that the oxidizing conditions
are due to high ultraviolet flux, because the
site is only 1 km above sea level. Instead, the
dry conditions in the Atacama must inhibit
biological production of reductants and pos-
sibly enhance the survival of photochemical-
ly produced oxidants. Our results suggest that
in the extreme arid core of the Atacama, we
have crossed the dry limit of microbial sur-
vival in extreme environments. The net result
is that photochemical processes dominate.
Thus, in the Atacama Desert, we find almost
no microorganisms and low levels of organic
material, and the organic material present
appears to have been oxidized. The LR ex-
periments confirm the presence of as-yet-
unidentified oxidants in the Atacama soil. In
many respects, these soils are similar to the
Mars soils investigated by the Viking Biolo-
gy Experiment and may provide a valuable
testing ground for instruments and experi-
ments designed for future Mars missions.
References and Notes
1. K. Biemann et al .. }. Geophys. Res. 30,4641 (1977).
2. The temperature of pyrolysis was set to any of three
temperatures: 200°, 350°, or 500°e. The detection
limit for benzene was less than 0.5 to 5 ppb; for
smaller molecules such as HC02H, the detection
limit was poor, at best in the ppm range.
3. V. I. Oyama, B. J. Berdahl,}. Geophys. Res. 82, 4669
(1977).
4. G. V. Levin, P. A. Straat,}. Geophys. Res. 82, 4663
(1977).
5. e. P. McKay et al., Planet. Space Sci. 46, 769 (1998).
6. A. Miller, in Climates of Central and South America,
W. Schwerdtfeger, Ed. (Elsevier, Amsterdam, 1976),
pp. 113-145.
7. M. T. K. Arroyo, F. A. Squeo, J. J. Armesto, e. Villagran,
Ann. Mo. Bot. Gard., 7S, 55 (1988).
8. e. P. McKay et al., AstrObiology 3, 293 (2003).
9. Rain and temperature data for Chile are available
from Direccion Meteorol6gica de Chile at www.
meteochile.cL Climatological data for Chile from
1912 to 1970 are available at http://docs.lib.
noaa.gov/rescue/data_rescue_chile.html.
10. Materials and methods are available as supporting
material on Science Online.
11. S. A. Benner, K. G. Devine, L. N. Matveeva, D. H.
Powell, Proc. Nat!. Acad. Sci. U.S.A. 97, 2425 (2000).
12. The organic concentration we see, albeit at much
higher temperature, is higher than the Viking pyr-GC-
MS reported limit.
13. R. e. Plumb, R. Tantayonon, M. Libby, W. W. Xu,
Nature 338,633 (1989).
14. We acknowledge support from NASA's Astrobiology
Science and Technology for Exploring Planets pro-
gram and Biomolecular Systems Research Program,
the National Autonomous University of Mexico
(grant nos. DGAPA-INl19999 and IN101903), the
National Council of Science and Technology of Mex-
ico (grant nos. 32531-T and F323-M9211), the
NASA-Ames/Louisiana State University Cooperative
Agreement (grant no. NCC 2-5469), the National
REPORTS
Science Foundation (award no. DEB 971427), and the
University of Antofagasta.
Supporting Online Material
www.sclencemag.org/cgilcontentlfull/302/564 7/10 1 81
DCl
Materials and Methods
SOM Text
Figs. Sl and S2
Data Tables Sl to S6
14 July 2003; accepted 29 September 2003
A Reservoir of Nitrate Beneath
Desert Soils
Michelle A. Walvoord,'* Fred M. Phillips,2 David A. Stonestrom,3
R. Dave Evans,4 Peter C. Hartsough,5,6 Brent D. Newman/
Robert G. Striegl'
A large reservoir of bioavailable nitrogen (up to ~104 kilograms of nitrogen per
hectare, as nitrate) has been previously overlooked in studies of globaL nitrogen
distribution. The reservoir has been accumulating in subsoiL zones of arid regions
throughout the HoLocene. Consideration of the subsoil reservoir raises esti-
mates of vadose-zone nitrogen inventories by 14 to 71 % for warm deserts and
arid shrublands worldwide and by 3 to 16% globally. Subsoil nitrate accumu-
Lation indicates Long-term leaching from desert soils, impelling further evalu-
ation of nutrient dynamics in xeric ecosystems, Evidence that subsoil accu-
mulations are readily mobilized raises concern about groundwater contami-
nation after land-use or climate change.
Increased deposition of bioavailable nitrogen
(N) at the land surface has adversely affected
water quality, biodiversity, and ecosystem
functioning around the world (1-6). Under-
standing such impacts requires quantification
of N sources, reservoirs, and cycling rates (J,
5, 7, 8). Desert soils, which cover approxi-
mately one-fourth of the conterminous Unit-
ed States and one-third of the land surface
worldwide, are reportedly low in total N (9,
10). Studies of N cycling in terrestrial eco-
systems have traditionally examined only the
biologically active soil zone, defined opera-
tionally as extending to ~ 1 m in depth (9,
11). Within this zone, N turnover is rapid (6),
and N concentrations decrease with depth (7,
10, 11), Natural sources of N in desert eco-
systems include nitrate (NO) -) and ammoni-
um (NH4 +) in precipitation, eolian deposition
of nitrate salts, and biological assimilation of
atmospheric Nz by N-fixing organisms (5, 7,
8, 10). Mechanisms of N removal include
'u.s. Geological Survey, Lakewood, CO B0225, USA.
2Department of Earth and Environmental Science, New
Mexico Institute of Mining and Technology, Socorro, NM
B7B01, USA. ·U.S. Geological Survey, Menlo Park, CA
94025, USA. ·School of Biological Sciences, Washington
State University, Pullman, WA 99164, USA. 5Graduate
Program of Hydrologic Sciences, University of Nevada,
Reno, Reno, NV 89557, USA. 6Desert Research Institute,
Reno, NV 89512, USA. 7Earth and Environmental Sci-
ence Division. Los Alamos National Laboratory, Los
Alamos, NM 87545, USA.
'To whom correspondence should be addressed. E-
mail: [e-mail]
plant uptake, volatilization to ammonium and
other gases, wind erosion, and denitrification
(6,7,12). Nitrogen loss from the soil zone by
leaching is generally assumed to be negligi-
ble in desert ecosystems (5, 10, 12). Our
findings challenge this assumption, demon-
strating that substantial quantities of N, as
NO) -, have leached and accumulated beneath
the soil zone over millennial time frames.
Soil-water N generally follows a nutrient-
type profile, with concentrations that decrease
sharply with depth because of biological uptake
and cycling (11). In contrast, soil-water chloride
(Cl-) follows a conservative solute-type profile,
with concentrations that increase with depth be-
cause of progressive evaporation and water ex-
traction by plant roots. In desert settings, Ct
typically exhibits an exaggerated conservative
solute-type profile resulting from the accumula-
tion of thousands of years of atmospheric Cl-
deposition (13). A recently developed model
(14) (supporting online material) quantitatively
explains these CI-profiles by considering geo-
theill1ally driven water vapor transport toward
the atmosphere, together with the hydraulic sink
created in the soil by the roots of desert plants,
Physical and biological processes selectively re-
move water, concentrating CI-(Fig. I A).
Surprisingly, soil-water concentration pro-
files of NO) -N in five arid-to-semiarid sites in
the western United States (Fig. 2) (15) follow the
conservative solute-accumulation profiles of Cl-
(Fig. 3) rather than the expected progressive
nutrient depletion profiles. Maximum NO)-N
o ...--o N
...--
N
L-Q) .c E
'*-Q)
Cf)
c o
~ q
Cl III E Q)
(,) c Q)
'(3
i
E .g
-0 Q) -0 III .Q c ~ o
www.sciencemag.org SCIENCE VOL 302 7 NOVEMBER 2003 1021
REPORTS
A
B
Atmosphere Precipitation & net dryfall
Soil
Subsoil
Evaporation Infiltration
Transpiration
V'\. Soil chloride pool ~O(v) '-------'
~~ 2 (1)~
Chloride
-leaching
Accumulating
subsoil chloride ~
reservoir H20(v)
Deep
profile
drying
(Net upward "', it: (Upward advection
< watervaporflux)-;-; 1---t-0fCI-) Net
(Net upward ~I II :_ (Downward > CI-flux -0
liqUId water flux) ,-t diffusion of CI,
Precipitation & net dryfall
Atmosphere
NH;
Soil , , , , , ,
NO;
I J~.
Nitrification ~
~NOi
...
" , :
; , ) Nitrogen fixation
" Organically bound nitrogen NP~·>N02·.J-· ' t.I -~. -Soil nitrate pool . • -' I ''2 -Assimilation : r · anima s ~
• Denitrification +_. NOi ' plants '----+ microorganisms
,,' 1--;-----' '" Nitrification ", & fungi --, 1 /·
, ,
,
, ,
I
NH humus
? 3
Subsoil~~ -
NH: 1--(Ammonification) +---j
(Leaching) _ Nitrate _ _ _ _ _ _ _ _ __
leaching y -----_ ......... --_ ...... -.. ,
N20 .........
N ...--NO; 1--Accumulating subsoil
2 nitrate reservoir
1 Scant : ______ i microorganisms, :
(Assimilation) 1 !~~~~, _ ~ _~':~':~ ___ : (Denitrification)
t~ (Upward advection of NOi) 'i> Net NOa-flux -0 I .. _(DOWnWard diffusion of NOi)
Fig. 1. M,ain transformations and transport pathways for water, CI-, and N03-in desert systems.
St raight lines denote liquid (I) pathways; wavy lines denote vapor (v) pathways; dotted lines denote
minor pathways. (A) Water and Cl-pathways. CI-arrives at the land surface in dust and
precipitation, accumulating near the land surface (soil CI-pool) during normal conditions of limited
rainfall separated by prolonged droughts, Conservative Cr anions are completely excluded from
soil vapor and preferentially excluded by ceiL membranes from plant transpiration. Soil CI-Ieaches
to the subsoil reservoir during Infrequent major wett.lng events. Water returns to the soil zone and
atmosphere as vapor, leaving nonvolatile CI-behind. Small net fluxes of water and CI-beneath the
upper subsoil are directed upward, reflecting deep-profile drying under current climatic conditions.
(8) NO]-pathways. N03-acts like CI-with respect to leaching and e)(clusion from soil vapor.
Unlike CI, however. N03 -Is preferentially taken up by plants and is reactive. Assimilation,
nitrification, denitrification. and ammonification are all biologically media ted. Subsoils, beneath the
root zone, are virtually devoid of organic matter and active organisms, where leaching and
evaporative concentration are the main processes affecting NO] -. Soil-pool flushing after extended
dry periods, when limited bioavailable carbon reserves are exhausted, leaches accumulated NO)-
to the subsoil below the reach of plants.
~ I Geographic boundaries :....J of U.S. deserts
Fig. 2. Map showing locations of vadose-zone
pore-water concentration profiles used in this
analysis.
concentrations in the subsoil below these nutri-
ent-limited vegetation communities (10) can ex-
ceed 2000 mg liter-J, surpassing N concentra-
tions applied in hydroponic agriculture by a fac-
tor of 10. Clearly, not all N03 -N is consumed in
the soil zone. We infer that N03-, like Cl-,
leaches from the soil pool to the subsoil reser-
voir, just beyond the reach of roots, during oc-
casional deep-wetting events. Once there, N03-
concentrates as water moves upward as vapor
along energy gradients and ultimately returns to
the atmosphere via plants (Fig. IB). The sus-
tained absence of downward water movement
below the subsoil reservoir has enabled N03-to
accumulate for thousands of years (13, 14).
Desert subsoils are persistently low in or-
ganic matter, low in microbial populations,
low in water content, aerobic, and neutral to
basic in pH (16); all of which promote N03-
stability and inhibit denitrification (17) .
Integration of the N03 -N profiles from
I m to the maximum depth sampled yields
subsoil N03 -N inventories that vary from 30
to 13,600 kg ofN ha-J (table S2). The N03-
N inventories show high intra-and interre-
gional variability relative to Cl-inventories.
This is not surprising, as N gains and losses
within the soil zone are controlled by uneven-
ly distributed plant and microbial activity, in
addition to hydraulic controls (Fig. I B). De-
spite the large variability, general trends are
apparent. For example, the pinyon-juniper
woodland in semiarid northern New Mexico
(Los Alamos) has the lowest N03 -N inven-
tory, suggesting a lower limit for environ-
mental conditions under which subsoil N03-
N accumulates in appreciable quantities. A
nearby ponderosa pine woodland that re-
ceives moderately more rainfall shows little
to no subsoil N03 -N accumulation (18).
One key factor contributing to contrasting
N03-behavior in arid and humid soils is the
establishment of a persistent hydraulic sink at the
base of the soil zone in deserts. Cl-mass balance
calculations provide an estimate of the time
scales over which conditions required for solute
accumulation have been maintained (13, 14)
(supporting online text). Estimated accumulation
times for the desert sites range from 10,000 to
1022 7 NOVEMBER 2003 VOL 302 SCIENCE www.sciencemag.org
o
'\""" o N
..-
N .... Q) ..c E ~
Q) en
c o
~ ~
Cl 03 E Q)
C,,) c Q) '0
i
E .g
"C Q)
"C 03 .Q c ?: o o
REPORTS
Mojave Desert Sonoran Desert Chlhuahuan Desert High Plains Los Alamos Mesa Fig. 3, Vadose-zone ni-
trate N and chloride
pore-water concentra-
tion profiles from loca-
tions indicated in Fig.
1. (No chloride data
are available for HP2.)
Note the change of
the Cl (bottom) scale
for CD1. HP1. LAM1.
and LAM 2 and the
change of the N03 -N
(upper) scale for MD1.
LAM1. and LAM2. Data
sources are as follows:
~'I ra l~-N (illY I,', Nlln'k-N (llIg I " N llI:ltc-N Illig L I. \llrak'.;' (fn~1 1.1) Nllral(,.N (m e I. I)
" [IUIII 211(HI tlJOlI III!IO ') ':;1111 Intll) I ':>1111 2(111) " 'i'II) II Ii III I "Iilll ~r jlltl " .,,11l) IOJI(l 1.""tllI ':lino " ~~ I '" 1,1-)i!! 1110
0
--CI
5 ---N03--N
~ 10
.§
-:9 15 c.
8 20 --CI ---CI ----CI --CI
NOT-N N03'-N -~ NOr,N ~ NOr-N
25
MOl SOl ( III I-H'I LAMI
I) (OliO 211(10 .lOtiO .[OtIO I) ,f)111) 2Ulll) JfIUI) I(J!U) I) l tJOIJ.['JO()hf)I)ONf}IHI I) I()(' ~IHt lOll 'IIUI (I :!Oilfl ..f(HIP (,flOI)
MDl (74). MD2 (29).
SDl and SD2 (30).
CDl and CD2 (37).
HPl (32). HP2 (27).
and LAM 1 and LAM2
(33).
Ch loride (Illg L-') Chkmdc (mg L"l Chloride Img L'l Chloride Img I;') Chloride (Illg L")
~ 10
~
-:; 15 c.
8 2U
25
NII<alc-" Im g L',
II '"11 I n"~1 I .... I!/I .'\(lfl'l ..
--CI
-'-NOr,N
MD2
30 +-----.--....--.---1
SD2
Nllr;)lc-N 11l1~ l.", NltTak-N (m~ L.I )
~/J 1(1 (,n XII IHII
--"0-(I) ---CI
-.-(2) -'-NOr-N
---CI
--.-NOr-N
( '112 1U'2 lAM2
() lor)!) 200t) :'111)(1 ,llIfHI ~I I"'H' 211011 1(),)1I ,IOOfl " IOUI! 20()1~ l(HH) .j(100
Chlori<k (mg I:')
20(1f1 ,lnOli (,000
Chloride fm!; l.-') Chloride (litH L") Chloride (lug l:'1
16,000 years (table S2). These accumulation
times are consistent with the hypothesis that the
onset of arid Holocene climatic conditions and
succession to xeric vegetation (19) triggered sub-
soil solute retention (13, 14, 17). Although the
presence of biochemical pathways makes NO)-
transport much more complex than CI-transport
(Fig. 1), the strong similarity of the two profiles
at many sites (Fig. 3) suggests that subsoil input
histories, transport behavior, and accumulation
times are similar. In the sampled environments,
inferred subsoil NO) -N retention times exceed
the 3000-year soil organic N pool maximum
retention time (9) by as much as a factor of 5.
Comparisons of subsoil inventories to soil
inventories for the sampled regions (7, 10, 12)
as well as for arid-to-semiarid soil inventories
worldwide (9) indicate that subsoil N (as NO)-)
inventories are similar in magnitude to total soil
N inventories (Fig. 4 and table S2). Based on
these comparisons, subsoil NO) -N likely ac-
counts for a preponderance of total vadose-zone
N (ground surface to water table) in nonriparian
arid environments. The ratios of subsoil NO)-
N to total vadose zone N are 44 to -92% for
the Mojave Desert, 41 to 81 % for the Sonoran
Desert, and 41 to 62% for the High Plains
region. Subsoil NO) -N accounts for ~4 to
20% of the total vadose zone N in the Chihua-
llUan Desert and <4% in dry forests similar to
the Los Alamos sites. Assuming that compara-
ble inventories (1 to 5 kg ha-1) exist in the
3 X 109 ha of Earth's warm deserts and arid
14
{'" 12
os ~ 10
...I<i
I Roose oi'slIb-soil N03--N inventory (>1 111)
X Regionnl Gv.mge soil·N inventory (0 -I m)
Global average
soll-N Inventory
(0-1 m)9
'0 8 x -x +-Iemperate steppe
(n = 666)
Fig, 4. Comparison of
subsoil N03 -N invento-
ries (table S2) with aver-
age regional and selected
global soil N inventories.
Multiple soil N regional
averages correspond to
average measurements
collected under different
species. Sources for soil N
data are as follows: Mo-
jave and Sonoran Deserts
(70); Chihuahuan Desert
(34); High Plains (35); and
CO Plateau (7. 36). Los
Alamos data are com-
pared with CO Plateau
data based on similar veg-
etation. proximity. and
lack of local soil N data.
....
c
<II 00 0
6 +-Iemperate dry forest
(n = 105)
~ +-warm-subtropical
X desert (n = 10)
,S 4 Z
X 2 I
X I
Regions
shrublands, subsoil NO) -N inventories
contain approximately 3 to 15 Pg of bio-
available N. This compares to global total
estimates of 21 Pg in desert soils and 95 Pg
in all soils (9). Consideration of subsoil
NO) -N thus raises estimated global inven-
tories of vadose-zone N by 14 to 71 % for
desert regions and 3 to 16% overall. Re-
ducing uncertainty in these estimates will
require substantial sampling efforts, given
the variability among measured profiles.
Nevertheless, the large amount of subsoil
N03 -N indicated by all of the sites war-
rants consideration in assessments of global
as well as arid-land N distributions.
The indicated leaching of soil NO) -N to the
subsoil reservoir affects long-tenn N cycling
calculations that derive fluxes from residuals.
For example, estimates of N lost to the atmo-
sphere by volatilization and denitrification will
o .-o N
..-N
L-Q)
.0
E
Q)
0.. Q) en
c a
~ a
Cl (1]
E Q) (J c Q)
'(3
i
E g
-c Q) -c
(1]
.Q c 3: a o
www.sciencemag.org SCIENCE VOL 302 7 NOVEMBER 2003 1023
REPORTS
be inflated by the amount of N03 -N leached to
the subsoil; the latter is substantial at some of the
sites investigated. Averaged over Cl--based ac-
cumulation times, long-term N03 -N soil losses
via leaching to the subsoil reservoir range from
3 X 10-3 106.8 X IOOkgofNha-' year-I
(table S2). For comparison, mean annual inor-
ganic N in wet deposition (N03 - N plus
NH4 + -N) ranges from 0.8 to 4 kg of N ha-'
year-I in the western half of the United States
(20). Our data do not permit precise generali-
zation of N03-N soil leaching to subsoil
reservoirs. Even so, N03-N soil leaching
clearly constitutes an appreciable fraction of
atmospheric N deposition over large areas.
Leaching ofN from arid soil zones is unex-
pected, given the N-limited nature of desert
ecosystems and the high nutrient utilization ef-
ficiency of xeric plants (lO, 21, 22), and cannot
be readily explained. The presence of large
quantities ofN03 -N sequestered below a depth
of I m demonstrates that not all of the available
N03 -N is consumed in the soil zone or re-
turned to the atmosphere. Ecologic implications
follow, given the strong linkages between nu-
trient cycling and plant community dynamics.
Recent studies show that desert plants do not
necessarily take up water and nutrients simul-
taneously (23). In addition, some species may
rely solely on available N at the soil surface
(24). Such behaviors may help explain the ap-
parent paradox of N03 -N leaching from soils
populated by N-Iirnited vegetation.
Subsoil N03 -reservoirs also have implica-
tions for groundwater quality, as their mobiliza-
tion may adversely affect public water supplies.
Drinking water exceeding the maximum con-
taminant level established by the U.S. Environ-
mental Protection Agency of 10 mg of N03 -N
litecl is associated with methaemoglobinaemia,
miscarriages, and non-Hodgkin's lymphoma (3,
25). Investigations in the 1970s reported large
amounts of subsoil N03-in southern California
(26) and central Nebraska (27) that could not be
attributed to agriculture or other human activi-
ties. Similarly, investigations of high N03 -lev-
els in Las Vegas Valley groundwater near irri-
gated fields ruled out fertilizer, livestock, and
septic systems as sources of pollution (28). Re-
cent studies indicate that subsoil N03 -reservoirs
are readily mobilized to groundwater when
desert land is converted to irrigation (29) (fig.
S3). Dam construction or changes in climate and
vegetation could likewise mobilize subsoil ni-
trate reservoirs, with local to regional effects.
References and Notes
1. J. N. Galloway er al., BioScience 53, 341 (2003).
2. W. B. Bowden, Biogeochemistry 2, 249 (1986).
3. L. W. Canter, Nitrates in Groundwater (Lewis, Boca
Raton, FL, 1997).
4. P. M. Vitousek et al., Ecol. App/. 7, 737 (1997).
5. L. R. Boring. W. T. Swank, J. B. Waide, G. S. Henderson,
Biogeochemistry 6, 119 (1988).
6. W. H. Schlesinger et al., Science 247, 1043 (1990).
7. R. D. Evans, J. R. Ehleringer, Oecologia 94, 314 (1993).
8. C. C. Cleveland et al., Global Biogeochem. Cycles 13,
623 (1999).
9. W. M. Post, J. Pastor, P. J. Zinke, A. G. Stagenberger,
Nature 317, 613 (1985).
10. N. E. West, J. J. Skuljins, Nitrogen in Desert Ecosys-
tems (Dowden, Stroudsburg, PA, 1978).
11. E. G. Jobagy, R. B. Jackson, Biogeochemistfy 53, 51 (2001).
12. W. T. Peterjohn, W. H. Schlesinger, Biogeochemistry
10,67 (1990).
13. F. M. Phillips, Soil Sci. Soc. Am. J. 58, 15 (1994).
14. M. A. Walvoord, M. A. Plummer, F. M. Phillips, A. V.
Wolfsberg, Water Resour. Res. 38, 1308 (2002).
15. Vadose-zone cores were collected without drilling
fluids. Individual sediment samples were analyzed
for water content. Soil leachate aliquots were an-
alyzed for Cl and NO, -using high-performance
liquid chromatography or ion chromatography. The
primary form of N in the aerated subsoil vadose
zone is NO, -. Data sources and site descriptions
are listed in table Sl.
16. C. C. Ainsworth, F. J. Brockman, P. M. Jardine, in
Vadose Zone Science and Technology Solutions, B. B.
Looney, R. W. Falta, Eds. (Batte lie, Columbus, OH,
2000), pp. 829-923.
17. P. Hartsough, S. W. Tyler, J. Sterling, M. A. Walvoord,
Geophys. Res. Lett. 28, 2955 (2001).
18. B. Newman, unpubliShed data.
19. T. R. Van Devender, R. S. Thompson, J. L. Betancourt,
in North America and Adjacent Oceans During the
Last Deglaciation, W. F. Ruddiman, H. E. Wright Jr.,
Eds. (Geological Society of America, Boulder, CO,
1987), pp. 323-3S2.
20. These values were obtained from http://nadp.sws.
uiuc.edul (National Atmospheric Deposition Pro-
gram/National Trends Network, 2001).
21. K. LaJtha, Biogeochemistry 4, 265 (1987).
22. J. R. Gutierrez, W. G. Whitford, Ecology 68, 2032 (1987).
23. R. L E. Gebauer, J. R. Ehleringer, Ecology 81,1415 (2000).
24. R. D. Evans,). R. Ehleringer, Oecologia 99, 233 (1994).
25. B. T. Nolan, J. D. Stoner, Environ. Sci. Techno/. 34,
1156 (2000).
26. J. M. Klein, W. L. Bradford, Distribution of Nitrate and
Related Nitrogen Species in the Unsaturated Zone,
Redlands and Vicinity, San Bernardino, California (U.S.
Geological Survey, Water Resources Investigations
Report 79-60, Menlo Park, CA, 1979).
27. J. S. Boyce, j. Muir, E. C. Seim, R. A. Olson, Farm Ranch
Home Q. 22, 2 (1976).
28. J. W. Hess, R. O. Patt, Nitrogen Contamination Sourc-
es in Shallow Ground Water, Las Vegas Area, Nevada
(Publication Number 32, Desert Research Institute,
Univ. of Nevada System, Las Vegas, NV, 1977).
29. D. A. Stonestrom et al., Estimates of Deep Percolation
Beneath Native Vegetation, Irrigated Fields, and the
Amargosa River Channel, Amargosa Desert, Nye
County, Nevada (U.S. Geological Survey, Open-File
Report 03-104, Menlo Park, CA, 2003).
30. R. C. Rice, R. S. Bowman, H. Bouwer, Groundwater 27,
813 (1989).
31. M. A. Walvoord, thesis, New Mexico Institute of
Mining and Technology, Socorro, NM (2002).
32. P. B. McMahon et aI., Water Movement Through Thick
Unsaturated Zones Overlying the Central High Plains
AqUifer, Southwestern Kansas, 2000-01 (U.S. Geolog-
ical Survey, Water Resources Investigations Report
03-4171, Denver, CO, 2003).
33. B. D. Newman, Vadose Zone Water Movement at Area
G, Los Alamos National Laboratory, TA-54: Interpre-
tations Based on Chloride and Stable Isotope Profiles
(Los Alamos National Laboratory, report LA-UR-96-
4682, Los Alamos, NM, 1996).
34. S. Schmidt, W. H. SchleSinger, unpublished data.
35. P. ). Zinke, A. G. Stangenberger, W. M. Post, W. R.
Emanuel, J. S. Olson, Worldwide Organic Soil Carbon
and Nitrogen Data (Oak Ridge National Laboratory,
report TM-88S7, Oak Ridge, TN, 1984).
36. R. D. Evans, unpublished data.
37. We thank J. Betancourt, B. T. Nolan, H. J. Smith, and
three anonymous reviewers for comments on earlier
drafts of this manuscript and K. Dennehy, P. McMahon,
S. Schmidt, and W. Schlesinger for providing unpub-
lished data. This material is based on work supported in
part by SAHRA (Sustainability of Semi-Arid Hydrology
and Riparian Areas) under the STC program of NSF,
agreement EAR-9876BOO, and by additional NSF fund-
ing. EAR-9614646 (S. W. Tyler) and EAR-9614509 (F.M.P.).
This investigation was performed While MAW. held a
National Research Council Research Assodateship Award
at the U.S. Geological Survey in Lakewood, CO.
Supporting Online Material
www.sclencemag.orglcgi/content/fuIV302/5647/1021/DCl
SaM Text
Figs. S1 to S3
Tables S1 and S2
References
5 May 2003; accepted 29 September 2003
African Droughts and Dust
Transport to the Caribbean:
Climate Change Implications
Joseph M. Prospero 1. and Peter J. Lamb2
Great quantities of African dust are carried over large areas of the Atlantic and
to the Caribbean during much of the year, Measurements made from 1965 to
1998 in Barbados trade winds show large interannual changes that are highly
anticorrelated with rainfall in the Soudano-Sahel, a region that has suffered
varying degrees of drought since 1970. Regression estimates based on long-
term rainfall data suggest that dust concentrations were sharply lower during
much of the 20th century before 1970, when rainfall was more normal. Because
of the great sensitivity of dust emissions to climate, future changes in climate
could result in large changes in emissions from African and other arid regions
that, in turn, could lead to impacts on climate over large areas,
Aerosols, including mineral dust, can affect
climate directly by scattering and absorbing
solar radiation and indirectly by modifying
cloud physical and radiative properties and
precipitation processes (1). Over large areas
of the Earth, the atmospheric aerosol compo-
sition is dominated by mineral dust. Dust
storms and dust plumes are the most promi-
nent, persistent, and widespread aerosol fea-
tures visible in satellite images (2). Dense
dust hazes often cover huge areas of the
Atlantic, Pacific, and Indian oceans down-
1024 7 NOVEMBER 2003 VOL 302 SCIENCE www.sciencemag.org
o ..-o N
..-
N ..... Q) .c E Q)
C. Q)
CJ)
c a
~ q
Cl !1l E ~ c Q)
'(3
i
E g
'0 Q)
'0 !1l a C 3: a o