Carbon Dioxide Variation in a Hardwood Forest Stream

Carbon Dioxide Variation in a Hardwood Forest Stream: An Integrative Measure of Whole Catchment Soil Respiration. Jeremy B. Jones, Jr. and Patrick J. Mulholland. Ecosystems (1998) 1: 183– 196

There are 7 common ions in freshwater: Cations: Ca+2, Mg+2, Na+1, K+1; Anions: Cl-1, SO 4 -2, HCO 3 -1 Sources of ions in freshwater From the atmosphere. The principal atmospheric source of ions is from sea spray. Accordingly, the ratios of the common ions in rain and snow reflect the ratios in seawater: Na and Cl are often the most abundant ions in rain. (The concentrations of ions in rain are of course much lower than the concentrations in seawater. ) From rock weathering. Watersheds contain soils and rock which consist of a wide variety of different minerals. Weathering of the soil and rock minerals produces runoff containing the common ions (and many other ions in lesser concentration). Some sample reactions which illustrate the general nature of weathering: Na-feldspar weathering 2 Na. Al. Si 3 O 8 + 2 CO 2 + 3 H 2 O <----> 4 Si. O 2 + Al 2 Si 2 O 5(OH)4 + 2 Na+1 + 2 HCO 3 -1 Calcite weathering H 2 CO 3 + Ca. CO 3 <----> Ca+2 + 2 HCO 3 -1 There are of course a host of other soil and rock minerals subject to similar weathering reactions. In general, weathering reactions can be characterized as a weak acid (carbonic acid) slowly dissolving basic minerals. Weathering reactions result in runoff containing dissolved ions, and commonly, undissolved particles of less soluble minerals such as clays.

Concentration patterns in surface water (the Gibbs 1970 "boomerang" plot). Atmospheric dominance Lakes which receive rainwater and snowmelt (without much influence by rock weathering) contain water with low concentrations of ions. Because of the source of atmospheric ions, from sea spray, chloride and sodium are relatively more abundant than the other ions. (The relative abundance of salts is not precisely the same as sea water because there is some fractionation in evaporation and transport of sea salts. ) In addition, rainwater is in equilibrium with atmospheric CO 2. In effect, rainwater is (approximately) a dilute solution of carbonic acid with an admixture of a small amount of sea salt. Rock Dominance Water more or less in equilibrium with the materials in the drainage basin is characterized by higher concentrations of ions and an increased significance of Ca, Mg, and bicarbonate ions. Many of the worlds great rivers discharge water which can be characterized as rock-dominated. Evaporation-Precipitation Evaporation and fractional precipitation represent the third mechanism identified by Gibbs. The world ocean is the end of this process. Some rivers and lakes have a similar chemistry if they occur in regions which are arid. The overall concentration of ions increases with evaporation until selected minerals begin to precipitate because their solubility product is exceeded. Because Ca. CO 3 is commonly the first mineral to precipitate, the relative concentration of these ions begins to decline with evaporation. With further evaporation, other minerals, such as gypsum (Ca. SO 4) will also precipitate. As a consequence, the relative concentrations of Na+, K+, and Cl- increase. Comment: Recent studies have produced data sets which are not consistent with Gibbs boomerang. For example, Kilham (1990, L&O, vol 35: 80 -83) presented data for African waters. For his data set, the "rainwater" arm of Gibbs boomerang is missing, possibly because of very rapid rock weathering or because of terrigenous dust contributing to precipitation chemistry. Most of the African samples are above 100 ppm total dissolved salts, and Ca and Cl are low. In contrast, Eilers and Brakke (1992, L&O, vol 37: 1335 -1337) report data for Oregon-Washington and other locations which are generally of very low overall ionic concentration, but with a range of Na/Ca ratios which span most of Gibbs plot. Eilers and Brakke argue that the Gibbs model is misleading. Gibbs reply (1992, L&O, vol 37: 1338 -1339) discusses Eilers comments and defends the original model as still the most suitable for "major" water bodies.

The Carbonate System The carbonate system is important for a number of reasons. Natural waters are buffered with respect to p. H mostly because of the content of inorganic carbon species. In turn, p. H is an important controlling variable for many important geochemical reactions (e. g. solubility of carbonates). Many important biochemical reactions, such as photosynthesis and respiration, interact with the p. H and the carbonate system. 1. The chemical species of interest. The carbonate species: CO 2, H 2 CO 3, HCO 3 -, CO 3 -2 Water and its ionization products: H 2 O, H+, OHThe other common ions: Na, K, Ca, Mg, Cl, SO 4 2. Some definitions. Carbonate alkalinity is defined as: A = [HCO 3 -] + 2[CO 3 -2] + [OH-] - [H+] (where [species] implies the concentration, in moles/liter, of the species in question. ) By convention, [H 2 CO 3*] = [H 2 CO 3] + [CO 2] Total inorganic carbon, CT = [H 2 CO 3*] + [HCO 3 -] + [CO 3 -2]

The carbonate system may be completely described by 6 equations: (1) [H+][HCO 3 -]/[H 2 CO 3*] = K 1 = 10 -6. 3 (2) [H+][CO 3 -2]/[HCO 3 -] = K 2 = 10 -10. 25 (1 st dissociation of carbonic acid) (2 nd dissociation of carbonic acid) (3) [H 2 CO 3*] = p. CO 2 x k. CO 2 (Henry’s law for CO 2) (p. CO 2=10 -3. 5 atm, KH = 10 -1. 5) (4) [H+][OH-] = Kw = 10 -14 (Dissociation of water) (5) [Ca+2][CO 3 -2] = KSo = 10 -8. 35 (Solubility of calcite) (6) [H+]+[Na+]+[K+]+2[Ca+2]+2[Mg+2] = [Cl-]+2[SO 4 -2]+[HCO 3 -]+2[CO 3 -2]+[OH-] (Requirement for electroneutrality)

The ionization fractions (top panel of figure) are defined algebraically as: [H 2 CO 3*] = CT x a 0; [HCO 3 -] = CT x a 1; [CO 3 -2] = CT x a 2 The sum of the a ’s: a 0 + a 1 + a 2 = 1 The numerical value of the a ’s are a function of p. H alone. They are defined as: a 0 = {1 + K 1/[H+] + K 1 K 2/[H+]2}-1 a 1 = { [H+]/K 1+ 1 + K 2/[H+]}-1 a 2 = {[H+]2/K 1 K 2 + [H+]/K 2 + 1}-1 Definitions of equivalence points a. Endpoint for an alkalinity titration: point x This endpoint is defined by the proton condition: [H+] = [HCO 3 -] + 2 [CO 3 -2] + [OH-] (Or, a pure solution of CO 2) b. Endpoint for "phenolphthalein" alkalinity titration: point y: [H 2 CO 3*] + [H+] = [CO 3 -2] + [OH-] (Or, a pure solution of Na. CO 3) c. Point z: 2[H 2 CO 3*] + [HCO 3 -] + [H+] = [OH-] (Or, a pure solution of Na 2 CO 3) Notice that for all three of these points (x, y, and z), the p. H at which the endpoint occurs is a function of CT. For example, the p. H for x decreases as CT increases.

Alkalinity is a measure of the ability of a water system to resist changes in p. H** when acid is added to water. A stream that has a high alkalinity is well buffered so that large inputs of acid (from acid rain for instance) can be made with little affect on the stream p. H. A stream that has a low alkalinity is poorly buffered and may undergo large, sudden drops in p. H in response to acid inputs. • Alkalinity is roughly equivalent to the imbalance of cations and anions: Alkalinity = 2[Ca 2+] + 2[Mg 2+] + [Na+] + [K+] + [NH 4+] - 2[SO 42 -] – [NO 3 -] – [Cl-] • The charge imbalance is “corrected” by changes in equilibrium in the DIC system. • Thus: Alkalinity = [HCO 3 -] + 2[CO 32 -] + [OH-] - [H+] • Alkalinity increases by processes that consume SO 42 -, NO 3 - or other anions, or that release DIC. • Alkalinity decreases by processes that consume cations or DIC. • Acid rain decreases alkalinity due to additions of H+ and SO 42 -.

An alternative definition of alkalinity The current tendency is to use "acid neutralizing capacity", or ANC, instead of "alkalinity" to describe the acid buffering capacity of a natural water. This usage is justified because of the method used to measure ANC: an acid titration to a particular p. H endpoint (but note the use of the Gran plot) An alternate definition of alkalinity (or ANC) can be developed from the requirement for electroneutrality. Beginning with the electroneutrality equation the terms may be rearranged to give: [HCO 3 -]+2[CO 3 -2]+ [OH-]-[H+ ]=[Na+]+[K+]+2[Ca+2]+2[Mg+2]-[Cl-]-2[SO 4 -2] Thus, Alkalinity (or ANC) can be defined as: Alk = [Na+] + [K+] + 2 [Ca+2] + 2[Mg+2] -[Cl-] -2[SO 4 -2] This version of alkalinity explains why it typically behaves as a "conservative" quantity. Samples may be collected, stored, and analyzed as much as 6 months later. The conceptual definition based on carbon species would seem to imply that alkalinity should be rather volatile. The alternative definition based on common ions also draws attention to conditions under which alkalinity would not be conservative. Some examples: Redox processes which involve important buffering species will alter alkalinity. Examples are: oxidation of ammonia to nitrate (nitrification), oxidation of sulfide or reduction of sulfate, oxidation or reduction of iron. Precipitation or dissolution of minerals involving participating species will alter alkalinity. For example, the precipitation or dissolution of calcium carbonate can alter alkalinity. Other participating species, such as organic acids, may influence ANC, and may not behave conservatively. Note also that CT and p. H do not behave conservatively. They must be measured as quickly as possible to obtain an accurate estimate of the makeup of a water sample at the time of collection.

Long-term decline in carbon dioxide supersaturation in rivers across the contiguous United States Jeremy B. Jones Jr. , Emily H. Stanley, Patrick J. Mulholland GEOPHYSICAL RESEARCH LETTERS, VOL. 30, NO. 10 2003 1. Introduction The concentration of CO 2 is an important property of aquatic ecosystems, reflecting both internal carbon dynamics and external biogeochemical processes in the terrestrial ecosystem [Cole et al. , 1994; Jones and Mulholland, 1998 a; Richey et al. , 2002]. CO 2 and dissolved inorganic carbon (DIC) concentrations in rivers and streams result from an interplay between inorganic carbon fixation via aquatic primary production, organic matter decomposition, import via ground waters, and exchange with the atmosphere [Hope et al. , 2001; Palmer et al. , 2001]…. . (nice discussion) 2. Methods 2. 1. Dataset 2. 2. Chemical Analyses

Table 1. Long-term Trends in the Partial Pressure of CO 2 (p. CO 2), Dissolved Inorganic Carbon (DIC), and Associated Chemical and Physical Parameters Across the Contiguous United States Variable Mean p. CO 2 (ppmv) Alkalinity (m. Eq l 1) Trend (y 1) 2109 2622 Trend standard error -78. 4 0. 9

4. Conclusions [16] The decline in p. CO 2, given the lack of change in alkalinity, eq. DIC, or eq. H+, points to a reduction or alteration in the quantity and/or quality of carbon import from terrestrial ecosystems. These changes could be caused by a myriad of factors but we suggest that either a decline in soil respiration or alteration in the nature of the hydrologic connections between terrestrial and aquatic ecosystems are the likely explanations. Baseflows (i. e. , groundwater fluxes to streams) have declined in the southeastern U. S. over the past 30 years [Lins and Slack, 1999], which could reduce the import of terrestrial carbon to streams [Jones and Mulholland, 1998 b]. However, baseflow increased in other parts of the country where we also observed CO 2 declines. Terrestrial production appears to have increased given trends in North America of increased plant growth [Myneni et al. , 1997], afforestation [Delcourt and Harris, 1980; Dixon et al. , 1994], and nitrogen deposition [Townsend et al. , 1996], which may lead to increased soil respiration and CO 2. More recent evidence, however, suggests that increased nitrogen deposition may lead to reduced soil CO 2; results from experimental additions of nitrogen to ….

…forest soils have demonstrated a reduction in the rate of organic matter decomposition [Berg and Matzner, 1997; Berg and Meentemeyer, 2002]. The reduction in riverine p. CO 2 may have also been caused by a decline in riparian and wetland habitat. Wetlands and nearstream environments can affect in-river metabolism and p. CO 2 by releasing organic matter and contributing water supersaturated in CO 2 from soil respiration. Indeed, respiration rates in riparian soils can exceed those in adjacent forest and cropland ecosystems [Tufekcioglu et al. , 2001]. High rates of primary production and soil respiration of wetlands and riparian zones, coupled with dramatic losses of these habitats over the past century [Mitsch and Gosselink, 1993] should have a strong and direct effect on stream p. CO 2. Regardless of the cause, the trend in p. CO 2 indicates that gaseous carbon losses from terrestrial ecosystems via aquatic pathways have likely declined across much of the contiguous U. S. and that ecosystem functioning significantly shifted during the latter part of the 20 th century.

Increase in the Export of Alkalinity from North America’s Largest River Peter A. Raymond and Jonathan J. Cole 4 JULY 2003 VOL 301 SCIENCE In terrestrial systems, there are two major processes that sequester atmospheric CO 2: Organic carbon produced during photosynthesis can be stored on land or exported in rivers, or terrestrial alkalinity (carbonate and bicarbonate ions) produced during chemical weathering in soils can be fluvially exported. The fluvial export of terrestrial alkalinity is also the major source of oceanic alkalinity and is a key regulator of the Ca. CO 3 saturation state of the oceans (1). During chemical weathering, the atmosphere provides the reservoir of CO 2 either directly, or indirectly through the respiration of plant-derived organic matter. The lithosphere converts some of this CO 2 to dissolved bicarbonate or carbonate through the weathering of the parent rock material, and the hydrologic cycle transports this dissolved inorganic carbon, as alkalinity, to rivers and ultimately to the ocean. Regional chemical weathering rates are controlled…. . ,

………Finally, these data demonstrate an increase in the export of alkalinity from the Mississippi watershed over the past five decades, which demands an increase in the supply of protons to mineral surfaces or an increase in the rate of chemical weathering due to warming. The acidity of rain has not increased in recent decades (38), and therefore the increase in proton delivery does not appear to be linked to acid rain. Potential mechanisms for an increase in proton delivery include an increase in atmospheric CO 2, an increase rainwater throughput, or an increase in plant and microbial production of CO 2 and organic acids in soils due to biological responses to increased rainfall and temperature. Plant productivity responses to increased temperature and rainfall are already documented for the United States (39, 40). Determining the relative contribution of these mechanisms, which will vary in importance regionally, is a critical component of future research. In particular, because nitrogen loading has also increased in the Mississippi watershed over the study period and is linked to agricultural practices (41, 42), studies must also determine the relationship between nitrogen loading and alkalinity export. Nitrogen loading could stimulate both soil respiration (43) and nitrification (44), the latter of which represents a source of acid to weather minerals that does not consume atmospheric CO 2 and was found to be responsible for 6% of the bicarbonate generation in a carbonate-rich, extensively farmed watershed (45).