The concentrations of the dissolved elements is analyzed for each sample using standard techniques (as you saw at Coweeta).  Weighted averaging is done on the concentrations - this is calculated as the total amount of an element in a series of samples, divided by the volume of water in those samples.

    Nutrient flux across ecosystem boundaries is calculated as:
                1) multiplying the measured concentration of an element in a sample by the volume of that sample, and
                2) multiplying the measured concentration of an element in streamwater by the volume flow of the stream during
                    the sampling period.  The average in streamwater is the mean of the sample at the beginning of the interval and
                    end of an interval.

    Precipitation chemistry at HB has been monitored continuously since 1963, the longest record of precipitation chemistry in the United States!

    Precipitation Chemistry - An Overview
    The chemical concentration of precipitation is dominated by H⁺ ions (nearly 70% of the cation strength). The remaining cations of importance are NH₄⁺, Ca⁺², Na⁺, Mg⁺², and K⁺, in this order.  For anions, sulfate (SO₄^-) is the most prevalent.  It is 2.5x as important as the next most prevalent anion, which is nitrate (NO₃^-).  As a consequence, a gross characterization of the precipitation chemistry at HB would be that it is a solution of nitric and sulfuric acids with a mean precipitation of 4.1.

    There is a surprisingly large amount of carbon dissolved in precipitation (1.3 mg/l on average).  Only a small amount is due to organic acids.  Most of this carbon comes in as hydrocarbons, aldehydes, and tannins/lignins.  These carbon compounds may turn out to be a significant input of fixed energy to the ecosystem that heretofore was not appreciated by ecologists.  On a yearly basis, this amounts to 31 Kg ha^-1 yr^-1.

    Elements get into precipitation from a variety of sources, including ocean spray, terrestrial dust, gaseous pollutants, and volcanic emissions.  It is important that the amounts of cations balance the amounts of anions, at least in terms of equivalents.  Any discrepancies will show up as a variation in pH from neutral.  Ecologists determine if their samples are in balance by determining the total equivalents of cations and anions, and using the measured pH to make up the difference with either H⁺ or OH^- ions.  If these add up, then the system is balanced.  If not, there may be some ions that are not being measured, and which account for the missing equivalents.

    Researchers at HB feel confident that most of the dust that comes into HB originates from outside the system.  With the exception of N and S, little elemental deposition occurs from dust for the other major elements.

    Calculation of Mass Balances of Chemicals at HB
    The general equation for calculating mass balances is to multiply the measured concentration of an element (mg/l) by the volume of precipitation incident on a watershed (l/ha).  This gives mg/ha.  Each weekly event is totaled, and summing the weekly events over the entire year gives mg ha^-1 yr^-1.

    In 1963-1964, 73 Kg of dissolved substances entered HB per ha (1,320 Eq/ha).  This was a drought year, and the number of Eq was low.  In the wettest year on record, there were 24% higher Eqs coming in.

    Throughfall and Stemflow
    During the growing season, only 87% of incident rainfall reaches the forest floor.  The rest is intercepted by the canopy, and evaporated off.  Probably only a very small fraction is absorbed, if at all.  What doesn't evaporate off runs down stems and leaves, processes which can drastically change the chemistry of the rain.  Stemflow accounts for about 5% of the water striking the ground beneath the canopy.  Certain elements become enriched, including K (about 91x), phosphorus (about 18x), magnesium (15x), and calcium (10x).  Many other substances also increase, whereas H⁺ ions decrease, due to exchange processes within the canopy.

    Carbon compounds increase in stemflow - from 31 to 52 Kg ha^-1 yr^-1.  This means that 21 Kg are leached per ha from the vegetation each season, a considerable carbon drain and flux that needs to be taken into consideration.

    Streamwater Chemistry
    Analysis of streamwater chemistry is important because this represents the flux of nutrients out of the ecosystem.  Thus, it becomes imperative to know the fluxes, and what causes variation in output rates.

    Although water comes into HB as a dilute acidic solution, it leaves as a relatively neutral solution with a pH near 5.0.  Ca⁺² and SO₄^-2 dominate the system.  The ionic strength (in mEq/l) is twice that of the incoming rain, caused mainly by a concentration effect due to evapotranspiration (this concentrates elements in the streamwater).  Using average annual evapotranspiration measurements of about 38%, this should concentrate the elements by a factor of 1.6X.  So how to explain the extra 0.4% concentration factor?  Most likely, weathering makes up the difference.

    Again, as in rain, there are significant amounts of dissolved organic carbon compounds in the streamwater.  Most concentrations lie between 0.3-2.0 mg/l.

    It is interesting to note that even though streamflows may vary by a factor of four, most element concentrations do not vary by more than a two-fold factor.  Biotic activity is important - during the growing season, nitrate levels are low due to plant uptake, and in the winter are higher.

    At HB, they have related streamflow to water chemistry.  Their model assumes two major compartments of water:
                                        1) dilute rainfall and snowfall, and
                                        2) more concentrated ground water.

    They assume that streamwater is some mixture of the two.  During flood melt, most of the water is due to the snow.  During drought, most of it is due to ground water.  Today, one can follow such changes by measuring the stable isotope ratios in the streamwater (something we'll discuss later in the course).  Ground water ratios of heavy to light H differ from rain water, where the ratios are highly temperature dependent.  This is also a way to determine what water source trees are using in ecosystems!

    Much of the chemical composition of the streamwater happens within the soil itself.  This shows that streamwater chemistry is closely coupled to the terrestrial component of the ecosystem, and that disturbances in the terrestrial area can have drastic and severe consequences.  The low variability of elemental concentration despite high variability in flow rates, suggests that this may be a characteristic of eastern deciduous forests.  A similar pattern has been found at Coweeta by Johnson and Swank (1973).  This suggests that downstream water quality is heavily dependent on the health and functioning of the surrounding forests.  When such areas are disturbed through logging or agriculture, streamwater chemistry changes significantly.

    One interesting trend that has emerged from the HB data is a constant, but significant increase in nitrate concentrations from 1963 through 1974 (from 0.92 to 3.14 mg/l).  Possibly, two freezing events in the 70s has resulted in larger exports of nitrate (there is evidence that freeze-thaw events stimulate nitrification, mobilizing nitrate into streamwaters.  Are there other possible explanations?  What about increasing amounts of nitrates in precipitation?  There is no clearcut evidence that precipitation inputs are related to streamwater outputs.  What about interactions with other elements?

    There is an inverse relationship between nitrate and sulfate in streamwater: nitrates peak in winter, when biological activity is minimal, whereas sulfates peak in the summer.  Perhaps excess sulfates are causing nitrates to leave the system. But sulfates have not increased in amounts over this period, so this seems unlikely.

    Since concentrations of most elements varies little with streamflow, then mass outputs can be easily calculated by simply knowing how streamflow varies throughout the year.  As streamflow increases, so does mass output for most elements.

    Annual Mass Balances
    In terms of mass outputs for dissolved substances, sulfate and silica dominate the output budgets.  If you calculate the budgets in terms of Eq, then silica is replaced by Ca.  Annually (1963-1974) streams put out 147 Kg/ha of dissolved substances (this includes organic carbon).  If you take out silica, the amount drops to 110 Kg/ha.  Approximately 8 Kg of carbon compounds are lost annually.

    In addition to losses of dissolved substances, a substantial amount of material leaves streams as particulate matter.  This is eroded material, particles that wash into the stream from outside, and dust that may blow in. The weir systems used in HB contain a large collecting area behind the streamflow gauge, and periodically, this area was cleaned out to estimate particulate loads.  Millipore filters were also used to determine very small sized particulates.  Annually, only about 33 Kg/ha are lost via streamwater, showing how tightly controlled such losses are from intact watersheds.  Interestingly, particulate export is higher by about 30% in the summer than winter, due to more biological activity, and ice and snow in the winter, which stabilize streamflows.  Particulate outputs are heavily dependent on streamflow, in opposition to dissolved matter, which is less sensitive.  Outputs range form 7 to 120 Kg/ha, dependent on streamflow patterns.  For example, over a four year period, it was shown that 86% of the export occurred in only 1.6% of the time, and used only 23% of the total water.  Nearly 16% was exported in 0.0025% of the time, using just 0.2% of the water!!

    What impacts does this have on elemental cycling?  Turns out that most elements leave in the dissolved phase, so are dependent mainly on streamflow.  However, the output of Fe and P are more closely linked to outputs of particulates, and therefore their fluxes are more dependent on the magnitude of streamflows during a year.  Overall, dissolved losses greatly exceed particulate losses by about a factor of 5 (150 Kg vs 30 Kg).

    Recent Trends at HB
    Dry deposition remains exceedlingly difficult to quantify, but recent estimates put the amount of dry sulfate deposition at around 37% of the total inputs (wet plus dry).

    Over the past thirty year interval, there has been a noticeable decline in the inputs of H⁺ ion, sulfate, and calcium in precipitation (40%, 37%, and 75%, respectively).  The decline in base cations (Ca, Mg, Na, and K) has also been found in other eastern forests, and in Europe.  Most of the decline has been due to Ca.  Nitrate concentrations were inching upwards until the mid 1970s, but now seem to be remaining flat.
 
    In the streamwater, Ca and sulfate show significant declines of 48% and 22% over this interval (up to 1993).  This seems due to reduced atmospheric inputs.  Some of these trends required over 18 years of data before they could be regressed and be a significant relationship (i.e., p < 0.05 for the regression).  This shows the importance of very long-term monitoring of environmental trends.

Anderson, D.H. and H.E. Hawkes. 1958. Relative mobility of the common elements in weathering of some schist and granite
        areas. Geochim. Cosmochim. Acta 14:204-210.

Bormann, F.H., G.E. Likens, and J.S. Eaton. 1969. Biotic regulation of particulate and solution losses from a forested
        ecosystem. Bioscience 19:600-610.

Carlisle, A., A.H.F. Brown, and E.J. White. 1967. The nutrient content of tree stem flow and ground flora litter and
        leachates in a sessile oak (Quercus petraea) woodland. J. Ecology 55:615-627.

Carroll, D. 1970. Rock Weathering. Plenum Press, NY, NY. 203 pp.

Cryer, R. 1976. The significance and variation of atmospheric nutrient inputs in a small catchment system. J. Hydrology
        29:121-137.

Eaton, J.S., G.E. Likens, and F.H. Bormann. 1973. Throughfall and stemflow chemistry in a northern hardwood forest.
        J. Ecology 61:495-508.

Eaton, J.S., G.E. Likens, and F.H. Bormann. 1978. The input of gaseous and particulate sulfur to a forest ecosystem.
        Tellus 30:546-551.

Hedin, L.O., L. Granat, G.E. Likens, T.A. Buishand, J.N. Galloway, T.J. Butler, and H. Rodhe. 1994. Steep declines
        in atmospheric base cations in regions of Europe and North America. Nature 367:351-354.

Johnson, P.L. and W.T. Swank. 1973. Studies of cation budgets in the southern Appalachians on four experimental
        watersheds with contrasting vegetation. Ecology 54:70-80.

Likens, G.E. and F.H. Bormann. 1995. Biogeochemistry of a Forested Ecosystem, 2nd Edition. Springer-Verlag, NY.
        159 pp.

Likens, G.E., E.S. Edgerton, and J.N. Galloway. 1983. The composition and deposition of organic carbon in precipitation.
        Tellus 35B:16-24.

Likens, P.C. 1994. Publications of the Hubbard Brook Ecosystem Study. Institute of Ecosystem Studies, Millbrook, NY.
        104 pp.