MODELING THE DYNAMIC GEOCHEMISTRY OF PRAIRIE POTHOLE WETLANDS

Spatial and temporal variation in soil and aqueous geochemistry are important landscape attributes that influence ecosystem processes, functions, and services, including the composition of biological communities.  Here we focus on the geochemistry of the Cottonwood Lakes (CWL) area of Stutsman County, North Dakota.  This study site has been the subject of intensive biologic, hydrologic, and geochemical research over many decades and is typical of the Prairie Pothole region of the north central U.S., which encompasses 5.3 million acres of wetlands.  We are investigating the underlying controls on the very steep geochemical gradients that exist within these wetland systems, which are highly sensitive to climate and cycle between flooded and dry conditions. Geochemistry of individual wetlands can vary dramatically over distances of hundreds of meters from low salinity, rainwater-derived compositions in upland areas, to high salinity calcium, magnesium sulfate dominated systems in adjacent lower elevation flow-through and groundwater discharge settings.  The key processes involve a balance among inputs of rainwater and groundwater, soil-water interaction along flow paths, evapo-transpiration, and biogeochemical processes.  In order to understand this complex suite of processes, we are modeling the system using both inverse and predictive geochemical models.  Inverse (mass balance) modeling using the USGS code NetPath allows the quantification of the mass transfers that determine CWL water chemistry.  Forward geochemical modeling allows prediction of the system behavior as major system variables change.   To model geochemical processes, we utilize over a decade of data on wetland aqueous chemistry, archived water samples from a time period corresponding to drought recovery (1994-1996), and newly collected elemental (including trace elements), isotopic, and mineralogical data on the chemistry of CLA waters, groundwaters, and soils.  The geologic substrate is heterogeneous glacial till with a component of Cretaceous organic and pyrite-rich shale.  Preliminary results indicate the oxidation of this pyrite or leaching of pyrite-derived gypsum during groundwater flow that yields isotopically light (d³⁴S ~-20‰) sulfate that drives the wetland waters to a sulfate-rich end member.  Concurrently, groundwater flow leaches calcium and magnesium from dolomite (as determined by XRD) in the aquifer sediment.  This interpretation is bolstered by statistical analysis of wetland chemistry that shows strong correlation among calcium, magnesium, sulfate, and lithium (lithium is strongly enriched in marine shale).  Within the wetlands, the sulfate is microbially reduced to sulfide, which shifts the sulfur isotope value of sulfate to heavier values (maximum +4‰), while also leading to carbonate precipitation.  The wetland water chemistry is further modified by evaporation, which causes the oxygen/deuterium isotopes of the water to evolve to heavier values (range for our data; d¹⁸O -10.7 to -2.3‰; dD -9.5 to -28.8‰).  Small amounts of deeper groundwater, identified by the presence of chloride ion, enters distal groundwater discharge wetlands along the flow path further modifying the wetland chemistry and adding potentially toxic arsenic and selenium.  Knowledge of solute chemistry over drought and deluge cycles due to evaporation and dilution is needed to place wetlands into the proper context of processes driving wetlands to facilitate effective ecological interpretations of biological data.