Learning Center
Plans & pricing Sign in
Sign Out

Kendall NADP talk 5.ppt - National Atmospheric Deposition Program


									   Use of stable isotopes for tracing sources of
    atmospheric nitrate to aquatic ecosystems

                    Carol Kendall and Emily Elliott
                       U.S. Geological Survey
                           Menlo Park, CA

Other collaborators: Beth Boyer (SUNY-Syracuse), Doug Burns (USGS-NY),
Rick Carlton (EPRI-CA), Greg Michalski (UC-San Diego), Tom Butler (IES-
NY), Scott Wankel (USGS-CA), Karen Harlin (NADP), Greg Lawrence (USGS-
NY), Mark Nilles (USGS-CO).
 Where does atmospheric nitrate come from?

  Natural atmospheric processes
  Agricultural emissions
  Power plant exhaust
  Vehicle emissions                                           NO3) for 2000
                            Nitrate concentrations in mg/L (as NO3) for 2000

Can we use nitrate
isotopes to determine
the main sources of
nitrate at NADP sites?
What are isotopes?

Isotopes are part of an element that have different
numbers of neutrons (but all their isotopes have
the same number of electrons and protons).

What are stable isotopes?

Stable isotopes are ones that are NOT radioactive.

Example: The element N has 7 electrons and 7 protons. It
can have various numbers of neutrons. The two stable
isotopes have 7 and 8 neutrons, respectively (14N, 15N).
How isotopes are used to trace sources of nitrate:

There are 2 stable nitrogen isotopes (15N,14N).
      ratios of 15N/14N are reported as 15N.
      (“ values”, in units of permil = ppt = ‰).

There are 3 stable oxygen isotopes (18O,17O,16O).
      ratios of 18O/16O are reported as 18O.
      ratios of 17O/16O are reported as 17O.

Therefore, if different sources of nitrate have different 15N,
18O, or 17O values, we can sometimes determine how
much nitrate comes from the different sources
Ranges of 18O and 15N values of nitrate from different sources

                                        Note: good 18O
                                        separation of
                                        atmospheric and
                                        microbial sources
Several examples of studies where nitrate
18O has been useful for determining
sources of nitrate in streamwater:
              Use of 18O of nitrate to quantify relative proportions
              of microbial vs atmospheric nitrate during snowmelt
                    in streams in small forested watersheds

     These data are from Loch Vale (CO) and show that <50% of the nitrate
     in streamwater is derived from the melting snowpack in 1995.

                           From Kendall (1998); Campbell et al. (2002)
Seasonal contributions of atmospheric NO3 (from overland flow)
          and sewage in urban streams in Austin TX

                                              These data show
                                              that most of the
                                              nitrate in these
                                              urban streams
                                              during storms is
                                              derived from

                                               (from Silva et al., 2002;
                                                        Kendall, 1998)
   Surface water nitrate in the Mississippi River Basin has higher
  18O and lower 15N values than nitrate in the San Joaquin Basin

                                           Mississippi River
                                           San Joaquin River
                                           NE USA rivers,
                                           Mayer et al. (2002)

atmospheric nitrate appears to be a significant source in urban and
      forested catchments, for small and large watersheds
   What is causing the large range of 18O and 15N
   values in precipitation?

Seasonal and
spatial variation
in sources of
Seasonal and
spatial variations
in atmospheric
Both? Other

What is the
current state
of knowledge?
The 18O of atmospheric nitrate appears to be slightly bimodel,
perhaps because of mixing of sources with different values
 Seasonal shifts in the 18O of atmospheric NO3 from 3
                     sites in the NE

                  Winter               Summer                 Winter

                                                  (modified from Williard, 1999)

The seasonal patterns may suggest temporal variability in sources.
Kendall (1998) speculated that the seasonality might be caused by the
relative contributions from power plant exhaust vs vehicle emissions.
15N values for NOx from vehicle and power plant

                                                    Power plant
                      Modified from Heaton (1990)
Can the isotopic composition of atmospheric nitrate be used to
determine whether the nitrate is derived from vehicle exhaust vs
                   power plant emissions?

Source of the         15N            18O              Δ17O
Vehicle exhaust     -2 to -11‰    +20 to +25‰ ???

Power plants       +6 to +13‰     +50 to +80‰ ???

Terrestrial                       -5 to +23‰ ???          0‰

Atmospheric                           >50‰          +14 to +30‰ ???
Relations between the 18O and 17O of terrestrial and
               atmospheric materials
  (caused by mass dependent and independent fractionations)

 Atmospheric NO3 is labeled by its 17O and 18O values

                                 Modified from Michalski et al. (2002)
 Origin of the high 18O and 17O values of atmospherically
                         derived NO3

NO generated by combustion, biomass burning, or biological emissions is
eventually oxidized to NO2 by several reactions. One such reaction is:

        NO + O3    NO2 + O2      (this O3 has high 18O and 17O values)

The NO2 is then photolyzed back to NO, and is the primary source of
troposphere O3:

        NO2 + hv  NO + O
        O + O2    O3

The NO is then re-oxidized to NO2 etc etc. Eventually NO2 is removed by
several “sink” reactions. For example:

        NO2 + OH  HNO3

Nitric acid can then react to form nitrate:

        HNO3 + NaCl  NaNO3 + HCl
   Updated ranges of 18O and 15N values of nitrate from
                    different sources

                                       high atmosphere sources ?

                                      low atmosphere sources ?

High nitrate 18O values are caused by the high ozone 18O values,
    and should be associated with high 17O and 17O values
      Conceptual model for how nitrate derived from car
    exhaust might have a different isotopic signature than
             nitrate derived from power plants.

Our hypothesis is that NOx from car exhaust gets most of its O from
photosynthetic O2 (with low 18O and 17O) whereas NOx from power plants
gets most of its O from tropospheric ozone (with high 18O and 17O).
  We will be testing our
  conceptual model with
  NADP samples

Alternative explanations for some of the 18O and 17O variability:

(1) The spatial variations could be caused by changes in atmospheric oxidation capacity
or changes in relative proportion of high atmosphere vs low atmosphere penetration of

(2) The seasonal variations could be caused by seasonal differences in reactions in the
upper atmospheres. The 18O of atmospheric NOx, regardless of source, might be
partially controlled by the relative contributions of the two primary NOx sink reactions.

During the winter when there is less daylight (hv), reactions that involve exchange with
O3 are the main source of nitrate, and the resulting 18O of NO3 is higher.

In the summer when photolysis produces maximum OH concentrations, reactions with
water vapor (where the oxygen in OH is -30 to +2 ‰) are a main source of nitrate, and
the resulting 18O of NO3 is lower.
Study design: analyze 2-month composites of archived
2000 samples from ~150 sites for nitrate isotopes
                                        Site selection
                                         All NY sites
                                        (because of our NY-
                                        funded study); all CO
                                        sites (because of
                                        USGS studies).
                                         All sites in
                                        National Parks,
                                        LTER sites, and in
                                        important USGS-
                                        studied watersheds.
                                         Tom Butler (IES)
                                        helped pick a
                                        reasonable spatial
                                        distribution of sites
                                        (for later atmospheric
                                        source models).
           NADP composite sampling

• Archived precipitation from 2000
• Samples composited over 6 2-month periods
• Composites volume-weighted within each 2-
  month period
• Samples with insufficient rainfall for complete
  chemical analysis not included in composite
• Analytical requirements for isotopic analysis:
  ~30 nmoles NO3-
                                 Analytical Methods
18O and 15N: Use the Sigman-Casciotti microbial denitrification
method to convert ~20 nanomole aliquots of nitrate quantitatively to

Analyze the N2O for 18O and 15N using an automated headspace
sampler on a continuous flow stable isotope mass spectrometer.

17O: We will convert a split of the N2O
to O2 and analyze this for 17O to
calculate 17O; this will also be used to
correct the 15N values for 17O effects.
                             Water Trap
                                 CO Trap
         Valves      GC Column

            Double                         Heating
            Needle                         Sleeve

            Sample Rack                    Liquid

      Flow and Temperature Controls
An important preliminary test: is the nitrate concentration in
archived 2000 samples affected by storage?
Another preliminary test: how well do our composites represent
the average concentration for the two months sampled?
First results for AL10:
 Our NY study:

         Quantifying Atmospheric Nitrogen Sources with
                 New Stable Isotope Techniques

Carol Kendall                Elizabeth W. Boyer                 Douglas A. Burns
U.S. Geological Survey       State University of New York       U.S. Geological Survey
WRD-Nat’l Research Program   Coll. of Envir. Sci. & Forestry,   WRD-NY District
Menlo Park, California       Syracuse, New York                 Troy, New York
Richard G. Carlton           Emily M. Elliott                   Greg Michalski
Electric Power Research      U.S. Geological Survey             University of California
   Institute                 WRD-Nat’l Research Program         Dept. of Chem. & Biochem.
Palo Alto, California        Menlo Park, California             La Jolla, California

     Other collaborators: Tom Butler (IES), Scott Wankel (USGS),
     Karen Harlin (NADP), Greg Lawrence (USGS), Mark Nilles (USGS).
Our study design is to assess spatial/seasonal isotopic variations in wet
precipitation at NADP sites and dry precipitation at selected sites, and see
if we can model the seasonal contributions of nitrate sources in a few
representative watersheds with good land-use data.
We also will analyze tree rings from selected locations in NY to
look for spatial changes in N sources

           15N of trees at different distances from a
                      roadway thru a forest

         From Saurer, M., Cherubini, P., et al. 2004,
         Submitted to Atmospheric Environment
FUTURE PLANS: We want to test the spatial and temporal
 predictions of mass balance models such as SPARROW
  with nitrate isotope data from large rivers (to compare
       with nitrate isotope data from precipitation).

      SPARROW predictions about
     sources of N in streams of major         Isotope data are
               watersheds                     particularly valuable
  Fertilizer               Atmospheric        because they can provide
                                              information on:
                                              (1) seasonality of source
                                              (2) comparison of input
  Wastewater             Animal agriculture   contributions versus
                                              export contributions.
                                              This information is critical
                                              for successful
                                              remediation of the actual
                                              critical loads.
               Smith and Alexander, 2000
Final philosophical point:

           Value of isotopes for water resources

    At a recent IAEA meeting, Andy Herczeg suggested that
    one main value of isotope techniques is:

       To tell us things about water resources that we didn’t
       know before.

    I propose that an even more important value is:

       To tell us things about water resources that
       CONTRADICT what we thought we knew before.

To top