This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.
Article pubs.acs.org/est
Using Continuous Underway Isotope Measurements To Map Water Residence Time in Hydrodynamically Complex Tidal Environments Bryan D. Downing,*,† Brian A. Bergamaschi,† Carol Kendall,‡ Tamara E. C. Kraus,† Kate J. Dennis,§ Jeffery A. Carter,§ and Travis S. Von Dessonneck† †
U.S. Geological Survey, Sacramento, California 95819, United States U.S. Geological Survey, Menlo Park, California 94025, United States § Picarro, Inc., Santa Clara, California 95054, United States ‡
S Supporting Information *
ABSTRACT: Stable isotopes present in water (δ2H, δ18O) have been used extensively to evaluate hydrological processes on the basis of parameters such as evaporation, precipitation, mixing, and residence time. In estuarine aquatic habitats, residence time (τ) is a major driver of biogeochemical processes, affecting trophic subsidies and conditions in fish-spawning habitats. But τ is highly variable in estuaries, owing to constant changes in river inflows, tides, wind, and water height, all of which combine to affect τ in unpredictable ways. It recently became feasible to measure δ2H and δ18O continuously, at a high sampling frequency (1 Hz), using diffusion sample introduction into a cavity ring-down spectrometer. To better understand the relationship of τ to biogeochemical processes in a dynamic estuarine system, we continuously measured δ2H and δ18O, nitrate and water quality parameters, on board a small, high-speed boat (5 to >10 m s−1) fitted with a hull-mounted underwater intake. We then calculated τ as is classically done using the isotopic signals of evaporation. The result was high-resolution (∼10 m) maps of residence time, nitrate, and other parameters that showed strong spatial gradients corresponding to geomorphic attributes of the different channels in the area. The mean measured value of τ was 30.5 d, with a range of 0−50 d. We used the measured spatial gradients in both τ and nitrate to calculate whole-ecosystem uptake rates, and the values ranged from 0.006 to 0.039 d−1. The capability to measure residence time over single tidal cycles in estuaries will be useful for evaluating and further understanding drivers of phytoplankton abundance, resolving differences attributable to mixing and water sources, explicitly calculating biogeochemical rates, and exploring the complex linkages among time-dependent biogeochemical processes in hydrodynamically complex environments such as estuaries.
■
and Delta.6,7 Further, studies suggest that τ affects the quality of fish-spawning habitats by affecting food supply and conditions for larval fish, with higher τ enhancing food production and accumulation.8,9 Evaluating the ecological effects of τ on ecosystem components (e.g., nutrients, phytoplankton, zooplankton, suspended particles) in dynamic systems is difficult due to the mismatch in methods for characterizing τ versus sampling for process studies. For example, in the Sacramento−San Joaquin Delta, τ has been estimated using simple box models, tracer models (e.g., salinity, dyes), and particle-tracking models.10 Such models often assume synthesized or idealized flow conditions and may miss ecologically significant low-flow subdomains and interactions with tidal wetlands. Nevertheless, particle tracking models have been used to compute τ to, in
INTRODUCTION
Water residence time (τ) is a master variable in aquatic systems. This parameter, an estimate of the average time a water molecule resides in a given area, influences concentrations of nutrients, oxygen, phytoplankton, zooplankton, suspended particles, and contaminants, thus impacting aquatic ecosystem function. Long residence times (e.g., high τ), for example, can promote the retention of exogenous nutrients and the development of phytoplankton blooms.1 High τ can also affect phytoplankton species composition by benefiting slow-growing species (e.g., cyanobacteria) that otherwise cannot outcompete faster-growing species.2,3 Changes in τ in the Sacramento−San Joaquin Delta and San Francisco Estuary are suspected to have led to an increased abundance of the potentially harmful cyanobacterium Microcystis aeruginosa at the expense of species more beneficial to Delta and Estuary food webs.3−5 A decline in phytoplankton suitable as a zooplankton food source is thought to be a major factor leading to precipitous declines in endangered endemic fish species in the San Francisco Estuary © 2016 American Chemical Society
Received: November 16, 2016 Accepted: November 23, 2016 Published: November 23, 2016 13387
DOI: 10.1021/acs.est.6b05745 Environ. Sci. Technol. 2016, 50, 13387−13396
Article
Environmental Science & Technology turn, quantify effects of changing nutrient loads and concentrations,11 estimate chlorophyll concentrations,12 and assess rates of primary production.13 One Delta study found that the models were confounded by ever-changing inflows and exports.10 However, the spatial resolution of particle tracking models is frequently insufficient to guide field sampling and may not accurately represent the ever-changing geomorphic and hydrodynamic complexity, resulting in low confidence in the τ associated with any particular sample. In lakes, stable isotopes in water (δ2H, δ18O), often referred to as “water isotopes”, are commonly used to evaluate τ by using the extent of evaporation (determined from the isotope measurements) in combination with local meteorological information (e.g., evaporation rate and humidity).14,15 Recently, new instrumentation (i.e., diffusive samplers coupled to laserbased cavity ring-down spectrometers, CRDSs) capable of measuring δ2H and δ18O continuously and in real time has become available.16,17 In this study, we sought to employ continuous high-frequency (1 Hz) water isotope measurements on board a high-speed boat to better assess the spatial variation in τ across a range of adjoining estuarine habitats. A high-speed boat is necessary to minimize any confounding effect of tides on the observed spatial distribution in τ. Further, we sought to assess if data from concurrent on-board high-frequency measurements of nitrate, chlorophyll (fCHLA), and other water quality parameters18,19 could be used together with the resulting maps of τ to inform our understand of biogeochemical processes and rates. The results provided the basis for empirical assessment of how τ is related to the distribution of biogeochemical processes in the study area as well as the means to make explicit calculations of biogeochemical rates by taking advantage of simultaneous gradients in constituent concentrations and τ, which was made possible by the high spatial resolution and precise temporal and spatial registration between τ and constituent measurements obtained using simultaneous on-board measurements.
Figure 1. Sacramento−San Joaquin River Delta. The red box highlights the Cache Slough Complex (CSC) study area. The CSC contains two flooded tidal wetlands, Liberty Island and Little Holland Tract, which are hydrologically connected to multiple breached sloughs such as (1) Shag Slough (west), terminating at the confluence of Lindsey and Cache Slough, (2) Prospect Slough (south), (3) Liberty Cut (northeast), (4) Stair Step (north), and (5) Yolo Bypass Toe Drain (east). The Sacramento Deep Water Ship Channel to the east is an intact, isolated dredged channel.
hydrologically bounded by the Sacramento River on the south, historical agricultural levees (“the Stair-Step levee”) to the north, Shag Slough on the west, Yolo Bypass Toe Drain, and the Sacramento Deep Water Ship Channel (DWSC) to the east. The DWSC is a dredged dead-end channel with little riparian vegetation and no wetlands. The westernmost channel, Shag Slough, is a straight channel but is hydrologically connected to Liberty Island via several breaches. Prospect Slough, the center channel, is surrounded by wetlands and riparian vegetation and is much more hydrologically complex, as it has a braided channel and exchanges with several wetland areas. Tides are mixed semi-diurnal, with a maximum spring tidal range of 1.5 m and a minimum neap tidal range of 0.5 m. Field Data Collection. Underway measurements were made from the 26-foot USGS R/V Mary Landsteiner.19 Data collection lasted for about 4 h (10:30−14:30 h PST), starting at a flood to ebb tide transition and continuing through the ebb. Sample water was continuously pumped from a pick-up tube located at the stern of the boat, at a fixed depth of 30 cm. The pick-up tube was located alongside the keel chine, well forward of the engine to permit continuous, bubble-free water flow. The inlet water was directed through a 178 μm in-line strainer, a three-stage de-bubbler (no filtration), and then three parallel flowpaths. The first flowpath was directed through a pleated membrane filter (Osmonics Memtrex, 25 cm length, 0.2 μm pore size; MNY921EGS; Osmonics, Inc.) and an open split interface into the Picarro continuous water sampler (CWS). The second flowpath was directed into a flow-through system consisting of a thermosalinograph (Seabird model SB45), nitrate analyzer (NO3−N mg/L; Satlantic model ISUS V3), and a YSI EXO2 fitted with sensors to measure conductivity, turbidity, pH, dissolved oxygen (DO), fDOM, and fCHLA. The third flowpath was used to compensate for changes in system pressure due to changes in boat speed. Further details about the
■
MATERIALS AND METHODS Study Area. The San Francisco Bay Delta (Delta) is the largest delta on the U.S. west coast, consisting principally of the confluence of the Sacramento and the San Joaquin Rivers (Figure 1). For more than a century, the Delta has been extensively altered to serve as the central element in California’s water supply system for agricultural demands (∼1 million hectares) and drinking water demands (∼27 million people) via diversion facilities designed to pump freshwater through the Central Valley Project and State Water Project conveyances. The Delta includes an assortment of tidal habitats affected by seasonally varying river inflows, local drainages, exports, and transport between the Delta and San Francisco Bay. Hydrologic and landscape-scale changes have channelized flows in the Delta, thus drastically altering τ and resulting in loss of habitat and biodiversity. Some native biota are now listed as threatened or endangered.9 This study was conducted in the Cache Slough Complex (CSC), located in the northern Delta (Figure 1). The CSC is geographically and hydrodynamically complex, containing areas of emergent vegetation, shoals, and sloughs, with tidal currents that mix and transport organisms and nutrients. The northern region consists of two tidal wetlands, Liberty Island and Little Holland Tract, that are hydrologically connected via multiple levee breaches and sloughs on the east (Prospect Slough and Liberty Cut) and west (Shag and Cache Sloughs). The CSC is 13388
DOI: 10.1021/acs.est.6b05745 Environ. Sci. Technol. 2016, 50, 13387−13396
Article
Environmental Science & Technology flow-through system and laboratory analyses are provided in the Supporting Information. Instrument data were displayed in real time so that the boat operators could slow down in areas of rapid change. The data acquisition system recorded data at 1 Hz, so at boat speeds between 5 and 10 m s−1, data points were
