Abstract:
Groundwater systems can im/mobilize phosphorus (P) and are an important source of P inputs to surface water bodies. Research on P pools and cycle in groundwater systems are therefore important for understanding a broader biogeochemical cycling of P. This thesis firstly identifies a methodology adapted to the groundwater environment for analyzing P pools. Secondly, the modified sequential extraction scheme was applied to analyze P cycling in the Ammer floodplain aquifer in Germany, addressing how total dissolved P (TDP) becomes enriched in the groundwater of a pristine calcareous aquifer. Thirdly, this thesis further explored the potential limitations and challenges of applying combination of stable oxygen isotope analysis of PO4 (δ18OPO4) with sequential extraction as analytical tool. The results showed that the extraction scheme derived from soil systems (Scheme A) performed better that the extraction scheme derived from marine systems (Scheme B). The expected surface-adsorbed inorganic P (PO4) pools by Scheme A and expected labile PO4 pool by Scheme B accounted for 49% to 89% vs 0% to 67% of the total extractable PO4 in tested synthetic minerals, and the extraction of organic P (Porg) by both methods were consistent with their effect on PO4 extraction. The understanding on roles of NaOH and CDB steps from extraction schemes were further enhanced, and the interpretation of their corresponding P pools in groundwater system has been corrected: NaOH-P should be interpreted alongside sediment mineral composition analysis to represent strongly surface-adsorbed P on Fe- and Al-minerals, moderately labile Porg, and relatively fragile structurally bound Fe-P; CDB-P should be interpreted alongside simultaneous measurements of Fe, Mn, and Ca to represent relatively strong structurally bound Fe-P, non-extractable surface-adsorbed P on Fe-minerals, and Ca-associated P. In the Ammer floodplain aquifer, TDP concentrations varied considerably from below the detection limit to 0.61 mg L-1 and were significantly associated with reducing redox conditions, dissolved organic carbon (DOC) and NH4+ concentrations, suggesting that the in-situ mineralization of organic matter (OM) was one of the main P-mobilizing processes. A pronounced proportion of surface-adsorbed PO4 and Porg (NaHCO3- and NaOH-extractable) in addition to PO4 that was structurally incorporated into Ca-minerals (HCl-extractable) were found in sediments, and Kd values of surface-adsorbed P suggested that Porg was found to be preferentially adsorbed over PO4 in aquifer material. The results suggest Ca-minerals and OM both represent the main P sinks in calcareous aquifer, while the mobilization of P via microbial mineralization of OM represents a comparatively fast process, and degradable OM and electron acceptors are both abundant in the calcareous aquifer of the Ammer floodplain. As a result, high concentrations of PO4 have subsequently accumulated in groundwater over time. The results of δ18OPO4 value in sequential extraction P pools indicated variability in the δ18OPO4 values across the different chemical extraction steps, for example, it implied that NaOH preferentially extracts isotopically heavier PO4 than other P pools within the same mineral, may therefore result in lighter δ18OPO4 value obtained in subsequent extraction steps. And I also observed the δ18OPO4 value in specific P pool of the mixture may vary from the δ18OPO4 value in P pool of corresponding single sample, when this P pool in the mixture is specifically contributed by corresponding single mineral therein. The variation of isotopic values in mixture might be attributed to two possibilities, i) preferentially extracted P between different combination of mineral types might be based on the isotope-related prioritization during chemical extraction which calls further study; ii) re-adsorption and re-precipitation of PO4 occurred during the extraction, which especially for the sample consists of Fe- and Ca-mineral. I suggest the differences in δ18OPO4 value among subsequently extracted P pools may be not necessarily due to pools were affected by the different P sources but rather to fractionation caused by the extraction schemes. However, due to lack of parameters regarding the properties of mineral and limited literatures, this observation calls for further research. This thesis further explores the research gaps in the study of P cycling in groundwater systems through methodological discussions involving synthesis, labeling experiments, and their applications to groundwater and aquifer sediments.
Phosphorus