The error in stream sediment phosphorus fractionation and sorption properties effected by drying pretreatments

Abstract

Stream sediment can control phosphorus (P) in the water column at baseflow. Two common laboratory analyses of sediment P are the equilibrium phosphate concentration at net zero sorption (EPC0) and P fractionation. Good sample handling ensures representative results, but oftentimes, studies rely on air-dried or freeze-dried samples, which alters sediment biogeochemistry. How and to what extent this influences EPC0 and P fractionation remains unclear. We therefore examine pretreatment effects on sediment EPC0 and P fractionation. We collected fine sediments (< 2 mm) from streams in the Tukituki River and Reporoa Basins in New Zealand (n = 31 sediments). Subsamples were then either kept fresh, frozen then lyophilized (freeze-dried), or dried at 40 °C for 2 weeks (air-dried). Measurements of EPC0 and P fractionation were made in triplicate. The sequential P fractionation scheme determined five different P pools: NH4Cl (labile P), NaOH reactive P (RP; metal oxide-bound P) and unreactive P (URP; organic P), HCl (Ca-mineral P), and residual P. Along with statistical comparisons between fresh results and the two pre-treatments, we explored correlations between pre-treatment effects and sediment physicochemical characteristics. The sediments had generally low EPC0 (majority < 0.020 mg P L−1), and uncertainty in EPC0 increased with concentration magnitude. While there were sediment-specific changes in EPC0 with pre-treatment, there was no consistent bias caused by pre-treatment. However, the differences between the fresh and air-dried sediment EPC0 were larger and more variable than between fresh and freeze-dried sediment. For P fractionation, the Tukituki sediments were enriched in HCl-P, while Reporoa sediments had more NaOH-RP and NaOH-URP. Despite large sediment-specific changes, the overall effects of freeze- and air-drying sediment were increased NH4Cl-P (estimated average effect, $$ \widehat{\theta} $$ = + 0.63 and + 3.7 mg P kg−1), no significant changes for NaOH-RP, contrasting changes in NaOH-URP (− 3.4 and + 3.3 mg P kg−1), and decreased HCl-P (− 40 and − 33 mg P kg−1). We found that drying sediment significantly influenced EPC0 and P fractions (especially the NH4Cl-P fraction). Air-drying was particularly error-prone and should be avoided. The use of freeze-drying to preserve samples for later analyses and improve ease of handling may be used with appropriate consideration of the research objectives and the error introduced by freeze-drying. However, we recommend using fresh sediments for analyses whenever possible, as they best represent natural conditions.