An Introduction to the Hawaii Ocean Time-series Data Organizational and Graphical System (HOT-DOGS)
Matthew J. Church (), Yoshimi M. Rii ()
Center for Microbial Oceanography: Research and Education (C-MORE)
Department of Oceanography, University of Hawaii, Honolulu, Hawaii
Introduction
The Hawaii Ocean Time-series (HOT) program is one of the world’s longest running, sustained ocean time series. The program began in 1988 with a near-monthly sampling schedule that continues today. HOT samples at Station ALOHA, a field site located approximately 100 km north of the Hawaiian island of Oahu in the North Pacific Subtropical Gyre (Figure 1). The time-series observations at Station ALOHA provide some of the only ship-based measurements from which to assess seasonal to decadal scale changes to an ocean ecosystem. While on station, the program utilizes a suite of oceanographic sampling equipment to collect seawater samples and measurements to assess time-dependent changes in physical and biogeochemical ocean properties and processes. The project relies extensively on a Conductivity-Temperature-Depth (CTD) rosette sampler (Figure 2) equipped with 24 twelve-liter sampling bottles: this piece of equipment provides high-resolution, real-time information on the vertical structure of several important ocean properties including temperature, salinity, oxygen, and fluorescence - an indicator of algal chlorophyll concentrations.
Measurements at Discrete Depths
The CTD rosette sampler also enables collection of seawater at 24 discrete depths between the surface ocean and the deep sea (>4500 m). Many of the measurements conducted by the HOT program require collecting a seawater sample. For example, HOT routinely measures the concentrations of the following:
· Various nutrients (nitrate, phosphate, silicate)
· Dissolved inorganic carbon (which includes CO2, bicarbonate, and carbonate)
· pH
· Various measurements of plankton biomass (particulate carbon and nitrogen, plankton cell counts, photosynthetic pigments)
All of these measurements require collection of a seawater sample and subsequent analyses of the dissolved or particulate matter constituents in the samples. In many cases, concentrations of dissolved and particulate material are in sufficiently low concentrations that the material needs to be concentrated prior to analyses. For particles, this is done by filtration, while concentration of dissolved constituents in seawater is often done by chemically precipitating the solute prior to determining its concentration.
Rate Measurements
In addition to the collection of samples using the CTD rosette, HOT conducts numerous rate measurements to quantify 1) how much carbon, nitrogen, and phosphorus are assimilated into plankton biomass on a daily basis, and 2) how much of that biomass is removed each day from the upper ocean by sinking of particles. Many of these rate measurements rely on free-drifting arrays; these arrays are used to conduct short-term experiments (usually 1-3 days) or to collect sinking particles.
Rates of Carbon Fixation
For example, HOT measures the daily rate of carbon fixation by photosynthetic plankton (a measure of primary production) by adding 14C-labeled bicarbonate to seawater samples and incubating the samples for the full daylight period on a free-drifting incubation array (Figure 3). These incubations provide a measure of the rate that the 14C was assimilated into plankton biomass. These samples are collected from and incubated at 6 discrete depths (5, 25, 45, 75, 100, 125 m); the incubation bottles containing the 14C labeled seawater are hung from a free-drifting array - this array is deployed before sunrise and recovered at sunset. The samples incubate at the same depths the samples were collected from-providing a way to mimic light and temperature conditions the plankton experience in the natural environment.
Rates of Sinking Particles
HOT also measures the rate that particles sink out of the upper ocean to the deep sea. This flux of plankton biomass from the upper ocean to the deep sea serves as the primary source of energy and food for deep-sea dwelling organisms. Moreover, the metabolic activity of these organisms living off this downward rain of material from above maintains vertical gradients in nutrients, inorganic carbon, and redox conditions in the sea. To quantify the flux (or rate) that particles sink to the deep sea, HOT utilizes sediment traps (Figure 4). These traps are made of cylindrical plexiglass and filled with a dense seawater brine solution. Each trap (12 in total) is deployed on a free-drifting array for ~3 days, collecting sinking particulate material. At the end of the collection period, the trap solution and collected particles are filtered and subsequently analyzed to determine the sinking fluxes of particulate carbon, nitrogen, phosphorus, and silica.
Zooplankton Biomass
To collect larger, rarer zooplankton, HOT relies on plankton nets (Figure 5). Mesozooplankton (weak swimmers in the 0.2-20 mm size range) are collected using vertically-integrating (0-175 m) tows of a 1 m2 net (202-µm mesh netting). The catch is then size fractionated by washing through a nested set of net filters and each fraction analyzed for wet and dry weight, and carbon and nitrogen concentrations. Tows are collected during the day and night to capture differences in zooplankton biomass due to diel vertical migrations.
Ocean Optical Properties
Light comprises a major structuring component to ocean habitats. When light strikes the surface of the ocean it can be absorbed (by the water itself or by material in the water, including plankton) or scattered (by particles in the water). The flux of light (irradiance) declines exponentially with depth in the ocean; the rate of exponential decay (termed attenuation) is determined by the scattering and absorptive properties of the seawater. Hence, in ocean waters where plankton are abundant, light is absorbed and scattered readily in the upper ocean, resulting in a decrease in the flux of light to depth. In oligotrophic waters, such as those found at Station ALOHA, low plankton biomass permits sunlight to penetrate deep into the upper ocean. However, understanding the factors controlling variability in the amount (flux) and quality (wavelength dependent attenuation) of light penetrating into the ocean provides important information on a major control on plankton growth. As a result, HOT measures a suite of optical characteristics of the ocean, including the depth of penetration of photosynthetically active radiation (PAR), and absorption and scattering of light at different wavelengths. Together, these measurements provide information on vertical and time-dependent changes in the optical characteristics of the seawater.
The Hawaii Ocean Time-series Data Organization and Graphical System (HOT-DOGS)
All of the HOT program data are publicly available through the internet. The main HOT website can be found at http://hahana.soest.hawaii.edu/hot/hot_jgofs.html; this website provides general information about the program, cruise schedules, methods and protocols employed by the program, and summarizes some of the major findings of the program. This website contains links to other useful sites related to Station ALOHA, notably including satellite imagery, provided by Dr. Ricardo Letelier of Oregon State University, of ocean color around the HOT study region. In addition, the website contains a link to the HOT program’s Data Organization and Graphical System (HOT DOGS): http://hahana.soest.hawaii.edu/hot/hot-dogs/interface.html. HOT DOGS is an online resource that can be used to access HOT data and/or create plots that depict vertical and time dependent structure in the NPSG ecosystem. The data system is an excellent teaching tool. Below, we provide description of how to use HOT-DOGS as a possible teaching tool, outlining several exercises that could be utilized to help students understand fundamental characteristics of the ocean habitat.
HOT-DOGS Modules
HOT-DOGS is organized around a series of modules that enable access to various forms of data. The front page of HOT-DOGS looks like this:
Along the top are a series of menu options that link to various modules in HOT-DOGS:
1. Data Extraction
2. Vertical Profiles
3. Horizontal Profiles
4. Miscellaneous
The Data Extraction module allows you to download multiple datasets at one time without a graphical interface; both the Vertical Profiles and Horizontal Profiles modules allow you to download single datasets (for example vertical profiles of nutrients or time series measurements of primary production) and provides a graphical interface that allows you to view the data. For teaching purposes, the most user-friendly modules are Vertical Profiles and Horizontal Profiles; the following exercises utilize these modules exclusively. Each pull down module menu choice provides two options: Vertical Profiles allows you to select “Display” or “Standard Intervals”, while the Horizontal Profiles allows you to select “Time-series” or “Contour”. Below we provide a few exercises that demonstrate the utility of each of these menu choices.
Vertical (Depth) Profiles
The Vertical Profiles module enables you to view and download vertical profiles of biogeochemical and physical oceanographic data collected at Station ALOHA.
Exercise 1: Temperature Profile
Let’s take a look at a temperature profile from the surface ocean to the sea bed:
1. Select Vertical Profiles à “Display” à “Bottle”
2. Select “CTD Temperature” from the pull-down menu for X-axis.
3. Select “Pressure” for Y-axis (note that oceanographers measure pressure as a means to determine the depth at which a property was measured. For example, in the surface ocean, a change in depth of 1 m is equivalent to a change in pressure of app. 1.007 decibars. For this purpose, you can consider pressure in decibars on the Y-axis as equal to depth in meters.
4. For data from Station ALOHA, leave the default “Station #” as 2. If you do not enter a cruise number or cast number, the default is to output all of the vertical profiles of temperature (measured by the CTD) collected since 1988. To view a plot of the data, leave the “output type” selected as GIF. When you hit the “Submit Query” link, you should see a plot that looks something like this:
This plot depicts a vertical profile of temperature from the surface ocean to the seabed (>4800 m). A few things to note:
1) Temperature decreases from the surface to depth, with the greatest depth-dependent changes occurring between ~100-500 m (this is zone of rapid change in temperature with depth is called the thermocline. The strong gradient in temperature in this region resists turbulent mixing, presenting a barrier to the movement of deep-sea water into the surface ocean.
2) Note that the spread of the data points is larger near the surface ocean and decreases with depth - this tells us that temperature is more variable in time in the upper ocean and much more stable in time in the deep sea.
If you want to have students practice manipulating data in a spreadsheet (Excel for example) you can download the data by selecting “Output Type” as “text”. This will provide a comma-separated text file of the data that can be readily imported into Excel. When you do so, note that you receive information on Julian day (abbreviated as “jday”), cruise number (abbreviated “crn”), Station (“stn”), “cast”, rosette bottle number (“ros”), the X-axis variable, and the Y-axis variable. Note that HOT DOGS uses October 1, 1988 as jday 1; hence, all dates listed as jday reference this start date (so jday 62 is October 1, 1988 + 62 days = December 2, 1988).
If we wanted to find out whether some of the variability in temperature is related to seasonal changes in the upper ocean, we would want to use the Horizontal Profiles module to examine how temperature varies over time at a selected depth.
Exercise 2a: Temperature at one depth over time (interannual to decadal)
1. Return to the main menu, select Horizontal Profiles à“Time series”à “Bottle”
2. Select “CTD temperature” as the Operator
3. Select “Depth” as the Y-axis
4. Change the “function” menu to “Horizon”, and enter “5” under “Value/Horizon”. This will output a time-series of temperatures measured at 5 m depth. When you submit the query you should see a plot that looks like this:
Several notable features can be observed in this plot:
1) In the near-surface ocean (5 m) temperatures at Station ALOHA vary ~3oC over an annual cycle.
2) There appears to be clear seasonal variation in temperature, with elevated temperatures in the summer, decreasing in the winter.
3) There appear to be interannual variations in temperature, perhaps driven by local weather or linked to climate fluctuations (El-Niño/La Niña oscillations for example).
HOT-DOGS will also allow you to bin the data to get a better sense of monthly, seasonal, or annual scale variability in temperature.
Exercise 2b: Temperature at one depth over time (monthly/seasonal)
Go back to the previous Bottle data display in Horizontal Profiles and select “Month” under “Grouping”; hit submit query and you should see a plot that looks like this:
Exercise 3: Do the same exercise to see whether there is any seasonal variation in temperature in the deep sea (select 4000 for the “Value/Range” option to examine temperature at a depth horizon of 4000 m). Note the scale on the Y-axis is rescaled for each selection, so although there may be some fluctuations in temperature, make sure to point out that the magnitude of that variation is very small (<0.1oC) compared to much larger fluctuations observed for the upper ocean.
Now let’s suppose you are interested in teaching students about nutrient cycles. You might start by asking students (without looking at HOT-DOGS) to make a plot of nutrient concentrations, e.g. nitrate, as a function of depth. Now use HOT-DOGS to demonstrate a few key points about nutrient cycling in the sea.
Exercise 4: Nitrate Profile
1. Select Vertical Profiles à “Bottle” à “Nitrate+Nitrite” as the X-axis. (Although Nitrate and Nitrite concentrations are reported together in HOT-DOGS (analytically it’s easier to measure them together in seawater), nitrite concentrations are a very small fraction of the total - most of the nitrogen in this determination is in the form of Nitrate.)
2. Select “Pressure” as the Y-axis, and leave the “station” as the default (“2”).
3. Hit submit query and you should see a plot that looks like this:
This plot depicts a vertical profile of nutrient concentrations in mmol/kg (for this purpose assume 1 L of seawater is equivalent to 1 L). There are several important features to highlight in the plot along with a couple of questions to discuss:
1) Nutrient concentrations are very low in the near-surface waters (<1 mmol/L), reflecting the rapid assimilation of inorganic nutrients into plankton biomass in the energy rich (=high light) region of the ocean.
2) Nutrient concentrations increase sharply below the well-lit upper ocean, with the largest gradient in nutrients observed within the thermocline waters. This reflects the increasingly important role for net remineralization of sinking plankton biomass below the well-lit region of the upper ocean. What types of organisms are responsible for maintaining this vertical gradient?