Metal forms and dynamics in urban stormwater runoff from a polluted creek system in Melbourne, Australia.

Published: 26 August 2021| Version 1 | DOI: 10.17632/n3xn3txy8w.1
Sarah McDonald


Stormwater runoff contains significant quantities of metal contaminants that enter urban waterways over short durations and represent a potential risk to water quality. The origin of metals within the catchment and processes that occur over the storm can control the partitioning of metals between a range of different forms. Understanding the fraction of metals present in a form that is potentially bioavailable to aquatic organisms is essential for environmental risk assessment. To help provide this information, the forms and dynamics of metal contaminants in an urban system were assessed across a storm. Temporal patterns in the concentration of metals in dissolved and total suspended solids (TSS) were assessed from water samples, and diffusive gradients in thin-films (DGTs) were deployed to measure the DGT-labile time-integrated metal concentration. Water samples were collected during baseline, during storm (inclusive of rising and falling storm limbs) and post storm conditions using grab sampling and rising stage sampler methods. Grab samples were collected as triplicates and rising stage samples were collected as duplicates. All DGTs were deployed in triplicates. All water samples were filtered (0.45 um) and sample filtrate was collected to assess the dissolved metal concentration, dissolved organic carbon concentration, dissolved organic matter characterisation and the concentration of major ions (including alkalinity). The 0.45 um filter was used to determine the TSS metal concentration. All sampling timepoints along the storm hydrograph were modified into stormtime by calculating the difference in hours (h) or days (d) of the sampling timepoints from the beginning of the storm (time = 0).


Steps to reproduce

The time averaged DGT-labile metal concentration (CDGT, ng/mL reported as µg/L) was calculated as described by (Zhang & Davison, 1995). Briefly, The measured metal concentration from DGT eluents were converted to mass of accumulated metal (M, in ng) on the binding layer using eq. 1 M= (C_e (V_e+ V_bl))/f_e (1) where Ce is the measured concentration of the metal in the eluent (µg/L), Ve and Vbl are the volumes (mL) of the eluent and binding layer respectively, and fe is the elution efficiency, which was 0.85 for all metals except Cr which was 0.8. The time averaged DGT-labile metal concentration (CDGT, ng/mL reported as µg/L) was then determined using eq. 2 C_DGT= (M∆g)/DAt (2) where M is the mass of the metal (ng) accumulated in the binding layer obtained from eq. 1, ∆g is the total thickness of the diffusion layer materials (0.094 cm, gel and filter membrane), D is the diffusion coefficient of the metal in the diffusion layer for the deployment temperature (in units of x10-6 cm2/s), t is the deployment time (s) and A is the surface area of the exposed filter membrane of the DGT device (3.14 cm2). All statistical analyses were performed using the statistical program R version 3.5.1.