Home
This Title All WIREs
WIREs RSS Feed
How to cite this WIREs title:
WIREs Water
Impact Factor: 4.451

Microlayer enrichment in natural treatment systems: linking the surface microlayer to urban water quality

Full article on Wiley Online Library:   HTML PDF

Can't access this content? Tell your librarian.

Natural treatment systems (NTS), such as constructed wetlands and stormwater ponds, are multibenefit, multidisciplinary approaches to sustaining water resources and reducing contaminant loading to urban streams. Surficial thin films (called surface microlayers) are not well characterized in NTS, but may have important implications for ecosystems, public health, and pollutant fate and transport. We present results from a case study evaluating microlayer contaminant partitioning across 4 NTS in Melbourne, Australia. To our knowledge, this study provides the first direct evidence for microlayer formation and contaminant enrichment (total petroleum hydrocarbons and trihalomethanes) in NTS. Contaminated microlayers were detected in the three most stable NTS, with stability defined relative to wind speed. Fluorescent‐dissolved organic matter profiles differed between microlayer and subsurface water in these systems, suggesting that fluorescence‐based techniques are useful for microlayer detection. Although individual fluorophores were not consistently associated with specific contaminants, fluorescence ratios were useful for identifying likely contaminant source waters, including road‐runoff and irrigation water from nearby green spaces. We evaluate our case study in light of what is known about surface microlayers in analogous systems (e.g., oceans, estuaries, and lakes), in order to identify existing research gaps and future opportunities. WIREs Water 2016, 3:269–281. doi: 10.1002/wat2.1128 This article is categorized under: Engineering Water > Sustainable Engineering of Water Engineering Water > Water, Health, and Sanitation Science of Water > Water Quality
(a) A schematic representation of surface microlayer structure inspired by figure 1(b) in Ref , showing intermixed organics (carbohydrates, proteins, and lipids), soil particles, nutrients, metals, and biotic communities (bacteria, viruses, phytoplankton, and protozoa/zooplankton). Although depicted as particles (red dots), the authors emphasize that microlayer‐associated nutrients and metals can be particulate, colloidal, dissolved, or some combination of the three. Note that microlayer constituents are not drawn to scale and are not proportional to one another. Although the microlayer shown here is approximately 400–1000 µm thick, microlayers can have a variety of thicknesses typically defined by the sampling method used (≤6 µm thick: polytetrafluoroethylene filter method to ≥1000 µm thick: original 1974 floating tray method). (b) The processes involved in microlayer formation: (1) bubble scavenging; (2) atmospheric deposition; (3) diffusion; (4) buoyant transport by TEP; (5) photodegradation [illustrated as solar‐induced splitting (dashed red line) and separation (opposing red arrows) of two protein subunits], (6) biodegredation and exudate release by phytoplankton or grazer feeding, (7) transport of motile organisms (and fecal pellets) into and out of the microlayer, and (8) other external biotic/abiotic processes such as foraging activity by birds and other wildlife, rainfall, and winds, which can effect microlayer composition and thickness.
[ Normal View | Magnified View ]
Cross plots of fulvic‐like (C1 peak) fluorescence intensity to (a) tryptophan‐like (T‐peak) or (b) tyrosine‐like (B peak) fluorescence intensity (arbitrary fluorescence units; a.f.u.). A black line shows the 1:1 fluorescence ratio in (a) and (b). Water samples from different NTS sites (this study) are marked by black circles and labeled with site names. Labels with ‐M and ‐B indicate samples from microlayer and bulk waters, respectively. Colored symbols show fluorescence ratios from samples (other studies) of various pollutants or water types; dark blue (clean rural river water), cyan (urban river water), light green (algae), gray (diesel fuel), purple (landfill leachate), brown (manure), yellow (recycled water), orange (secondary treated waste water effluent), pink (combined sewer overflow water), red (raw sewage). Where provided, error bars show the standard deviation in fluorescence intensity. Symbol shape relates specific fluorescence ratios to their respective studies. These are defined in the legend using the first letter of the first authors last name, and the study year; B‐05, B‐08, N‐01, BS‐04, B‐03, N‐07,, N‐05,, Q‐12, A‐08, Y‐12, B‐02, H‐10, H‐07, and H‐09. Most samples (this study and others) were filtered or centrifuged to remove particulates prior to analysis (exceptions: N‐01, B‐02, H‐10, A‐08, and Y‐12). Most samples (this study and others) were normalized to the Raman Intensity of a DI water blank (Intensity: 19 ± 1 a.f.u.). Exceptions include A‐08 (hexane blank required due to diesel hydrophobicity) and N‐05 (Super‐Q water + KCl blank).
[ Normal View | Magnified View ]
Percent fluorescence at each LID site for the five fluorophores identified by the best‐fit PARAFAC model (see Figure ). Protein‐like fluorophores (tyrosine‐like B peak and tryptophan‐like T peak) are shown in hot tones (reds). Humic‐like fluorophores (UV humic‐like A peak and visible humic‐like C2 peak) and fulvic‐like fluorophores (C1 peak) are shown in cool tones (blue and white). Microlayer samples are at the top of the figure and bulk water samples are at the bottom. The acronyms above each pie diagram correspond to the stations in Figure (BAN: Banyan Creek, OP: Ornamental Pond, NP: North Pond, and HRW: Huntingdale Road Wetland). Note that BAN is the only site with identical fluorescent signatures in microlayer and bulk waters.
[ Normal View | Magnified View ]
Best‐fit PARAFAC model with five components. (a–e) Excitation–emission spectra of model components; x‐axis is excitation (nm) and y‐axis is emission (nm). The spectra in panel (a) is consistent with UV humic‐like fluorescence (e.g., A peak), panel (b) is consistent with visible fulvic‐like fluorescence (e.g., C1 peak), panel (c) is consistent with visible humic‐like fluorescence (e.g., C2 peak), and panels (d) and (e) are consistent with tyrosine and tryptophan‐like protein fluorescence (e.g., B and T peaks, respectfully). (f–j) Excitation (solid line) and emission (dashed line) loadings of the components shown in panels (a–e).
[ Normal View | Magnified View ]
(a) Study area map; Melbourne, Australia. Red dots mark NTS sampling sites (BAN: Banyan Creek, OP: Ornamental Pond, NP: North Pond, and HRW: Huntingdale Road Wetland). White dots mark weather stations where rainfall and wind speed data were collected for BAN (ERD and FAWS) or HRW, NP, and OP (MA). (b) Cumulative rainfall (mm) at each NTS site for the period during sampling (red), 12 h prior to sampling (black), or 24 h prior to sampling (white). Note that rainfall is higher at OP and NP than BAN and HRW. (c) Average wind speed (km/h) at each LID site. Color coding is the same as in panel (b).
[ Normal View | Magnified View ]

Browse by Topic

Science of Water > Water Quality
Engineering Water > Sustainable Engineering of Water
Engineering Water > Water, Health, and Sanitation

Access to this WIREs title is by subscription only.

Recommend to Your
Librarian Now!

The latest WIREs articles in your inbox

Sign Up for Article Alerts